ORDER GRANTING PLAINTIFFS’ MOTION FOR SUMMARY JUDGMENT AND PERMANENT INJUNCTION

ROTHSTEIN, Chief Judge.

THIS MATTER comes before the court on a motion for summary judgment and a permanent injunction brought by Friends of the Earth and other environmental organizations (“FOE”) under their complaint for declaratory and injunctive relief. The instant motion concerns only defendant Army Corps of Engineers (“Corps”) and defendant the United States Navy (“Navy”). FOE asks the court to find that the environmental impact statements (“EIS”) prepared by the Navy and the Corps violated the environmental disclosure and informed decision-making requirements of the National Environmental Policy Act (“NEPA”), 42 U.S.C. § 4321 et seq., and that the Corps erred as a matter of law in granting the Navy a dredge and fill permit under Section 404 (“404 permit”) of the Clean Water Act (“CWA”), 33 U.S.C. § 1344. FOE seeks an injunction setting aside the Navy’s plans to dredge sediments in Everett Harbor and deposit them in Port Gardner Bay until such time as the Navy and the Corps adhere to their obligations under the environmental laws of the United States. Pursuant to the government’s request that the court announce its decision prior to July 22, 1988, the court issues this brief order, with a full memorandum decision to follow as soon as possible. Consequently, having reviewed the motion, together with all documents filed in support and in opposition, having heard oral argument by all parties, and being fully advised, the court finds and rules as follows:

Under the Administrative Procedures Act, which governs judicial review of an agency’s preparation of an EIS and of the Corps’ decision to grant a 404 permit, the court “shall set aside any agency action” undertaken “without observance of procedure required by law” or found to be “arbitrary, capricious, an abuse of discre *912 tion, or otherwise not in accordance with law." 5 U.S.C. § 706(2)(A), (D) (1982); Northwest Coalition for Alternatives to Pesticides v. Lyng, 844 F.2d 588, 590-91 (9th Cir.1988) (“NCAP”); Friends of the Earth v. Hintz, 800 F.2d 822, 831 (9th Cir.1986). The court recognizes that “NEPA is essentially a procedural statute” and that the court can not “substitute its judgment for that of the agency concerning the wisdom or prudence of a proposed action.” NCAP, 844 F.2d at 590-91. Likewise, the court also recognizes that it cannot set aside agency action as arbitrary or capricious “unless there is no rational basis for the action.” Hintz, 800 F.2d at 831.

The court has no doubt that both the Corps and the Navy made substantial efforts to meet their NEPA and CWA obligations. Nevertheless, the court concludes that, for a variety of reasons to be detailed in the memorandum decision to follow, the various EISs did not satisfy their two primary purposes: (1) providing decisionmakers with “an environmental disclosure sufficiently detailed to aid in the substantive decision whether to proceed with the project in light of the environmental consequences,” and (2) providing “the public with information and an opportunity to participate in gathering information.” Methow Valley Citizens Council v. Regional Forester, 833 F.2d 810, 814 (9th Cir.1987). Although the Corps issued a detailed Record of Decision (“ROD”) in support of its decision to issue a 404 permit, the court concludes that the ROD did not eliminate the areas of environmental concern; therefore, the Ninth Circuit’s recent decision in Half Moon Bay Fishermans’ Marketing A’ssn v. Carlucci, 847 F.2d 1389 (9th Cir.1988) is not apposite. In addition, because the Corps’ decision to grant the Navy a 404 permit relies heavily on the conclusions of the EISs, the rational basis underlying that decision no longer exists. Because of this, and because of various other procedural violations, the court concludes that the Corps’ decision to issue a 404 permit was both arbitrary and capricious and otherwise not in accordance with law.

A grant of summary judgment is appropriate here because no genuine issues of material fact exist with respect to the Corps’ and the Navy’s obligations under NEPA and the CWA, and because FOE is entitled to judgment as a matter of law. See IW. Electrical Service, Inc. v. Pacific Electrical Contractors Association, 809 F.2d 626, 630-631 (9th Cir.1987). Furthermore, for the reasons to be detailed in the memorandum decision to follow, the court concludes that an injunction should issue in these circumstances. The CWA’s dredge and fill permit requirements necessarily and inescapably require the issuance of an injunction. In any event, the NEPA violations are of such a nature that under traditional equity balancing the potential likelihood of environmental harm outweighs the national security concerns inherent in a delay of the Navy’s homeport project. See Amoco Prod. Co. v. Village of Gambell, Alaska, 480 U.S. 531, 107 S.Ct. 1396, 1402-04, 94 L.Ed.2d 542 (1987); Save the Yaak Committee v. Block, 840 F.2d 714, 722 (9th Cir.1988).

IT IS NOW, THEREFORE, ordered as follows:

Plaintiff FOE’s motion for summary judgment and a permanent injunction is GRANTED. The court SETS ASIDE the Navy’s 404 permit and enjoins the Navy’s dredging project until such time as the Navy and the Corps comply with NEPA and the CWA as explained in the memorandum decision to follow.

IT IS SO ORDERED.

ORDER GRANTING PLAINTIFFS’ MOTION TO CLARIFY INJUNCTION

THIS MATTER comes before the court on a motion by plaintiffs Friends of the Earth and other environmental groups (“FOE”) to clarify the court's previously issued injunction. Having reviewed the motions, together with all documents filed in support and in opposition, and being fully advised, the court finds and rules as follows:

*913 I

BACKGROUND

In a brief order issued pending publication of a more complete memorandum opinion, the court granted FOE’s motion for summary judgment and permanent injunction. Friends of the Earth v. Hall, NO. C88-380R, Order Granting Plaintiffs’ Motion for Summary Judgment and Permanent Injunction (July 20,1988). In framing injunctive relief the court stated that it “SETS ASIDE the Navy’s 404 permit and enjoins the Navy’s project until the Navy and the Corps comply with NEPA and the CWA ...” Id., at 912. Based on its understanding of the court’s intention, the Army Corps of Engineers (“Corps”) “suspended” the Navy’s permit and all work authorized under it, but allowed the Navy to continue with upland construction on the Homeport. Notice of Suspension of Navy’s Permit 2 (July 28, 1988) (“Suspension Notice”). Originally, FOE merely filed a response to the Suspension Notice. The government, however, treated FOE’s response as a motion to clarify or amend the court’s injunction, and filed a response. FOE then filed a proper motion to clarify, accompanied by a motion to shorten time and a reply to the government’s response. Because the government has already had an opportunity to respond, and has done so, and because FOE filed a reply brief, the court concludes that it is appropriate to rule on FOE’s motion at this time.

DISCUSSION

As explained above, after suspending the Navy’s permit and all work authorized under it, the Corps allowed the Navy to continue with upland construction on the Homeport, including demolition and construction over several hundred acres. Suspension Notice, at 2. According to the Navy, termination of upland construction work now underway will result in additional costs of $13 to $18 million. FOE asks the court to clarify the scope of the injunction, by expressly enjoining the entire Homeport project until the Navy and the Corps fully address all environmental concerns surrounding the project. FOE argues that the National Environmental Policy Act (“NEPA”), the National Defense Authorization Act of 1987, Pub.L. No. 99-661, § 2207 (1986) (“1987 NDAA) (copy attached as exh. 1, Plaintiffs Response to Notice of Suspension), and the National Defense Authorization Act of 1988, Pub.L. No. 100-180, § 2322 (1987) (“1988 NDAA”) (copy attached as exh. 2, Plaintiffs Response to Notice of Suspension) require the cessation of all Homeport activity.

The essence of FOE’s NEPA argument is that the upland work and the dredging activities are parts of one integrated project, and in order not to impair the government’s ability to perform balanced, thorough, environmental analysis and unbiased decisionmaking, NEPA requires that the injunction halt the entire project because the supplemental EIS would become “merely a progress report.... if any work continued during the review of the impact statement.” Stop H-S Assoc, v. Volpe, 349 F.Supp. 1047, 1049 D.Haw.), amended, 353 F.Supp. 14 (D.Haw.1972). FOE cites additional authority for this proposition. Massachusetts v. Watt, 716 F.2d 946, 952-53 (1st Cir.1983) (barred sale of offshore leases while new EIS prepared because otherwise the new EIS “may bring about a new decision, but that it is much less likely to bring about a different one”) (emphasis in original); Highland Cooperative v. City of Lansing, 492 F.Supp. 1372, 1383 (W.D. Mich.1980) (enjoining all parts of highway project because “[t]o allow defendants to make a substantial commitment of resources to one segment, while preparing the EIS for the other would be to avoid the spirit and letter of NEPA”). As the First Circuit has recognized, the risk of bias resulting from the commitment of resources prior to a required thorough environmental review is the type of irreparable harm that results from a NEPA violation. Massachusetts v. Watt, 716 F.2d at 952-53.

Although the court has real concern that continued upland construction would in fact impair the Corps’ ability to engage in unbiased decisionmaking, the court finds it un *914 necessary to rule on whether NEPA requires the court to enjoin the entire Home-port project. Rather, the court finds that the NDAA unequivocally prohibits the expenditure of any Homeport funds until the Navy and Corps address the environmental concerns set out in the court’s forthcoming memorandum opinion.

Homeport appropriations for fiscal years 1987, 1988, and 1989, contain a provision prohibiting the Navy from obligating or expending funds for the Homeport until “all Federal, state, and local permits required for the dredging activities to be carried out with respect to [the Homeport] have been issued.” 1987 NDAA, § 2207 (a) (1); 1988 NDAA, § 2322 (c). The legislative history clearly demonstrates that Congress intended to prohibit expenditures of Homeport funds “to ensure that environmental concerns are fully addressed prior to the initiation of construction.” Conf. Rep.H.R. 1748, 1987 NDAA (attached as exh. 3, Plaintiffs’ Response to Suspension). Based on these provisions, the Ninth Circuit held that the NDAA required an injunction covering all Homeport expenditures until the Washington Shorelines Hearings Board issued its decision on the Navy’s shorelines permit. Friends of the Earth v. Navy, 841 F.2d 927, 932 (9th Cir.1988).

The government raises two arguments for why the court should not craft its injunction to conform with Congress’ intent. First, in a hypertechnical procedural argument, the government contends that the court should not even address the NDAA issue because FOE brought this lawsuit to enforce NEPA and the Clean Water Act (“CWA”), not the NDAA. Because FOE’s complaint asked the court to enjoin activities covered by the 404 permit, Compl., at 26, the government contends that FOE must file a new action for the court to address the NDAA claim. However, FOE asked the court to declare defendants’ actions as “arbitrary and capricious, not in accordance with law, contrary to 5 U.S.C. § 706 (2)(A).” Compl. 115. As to relief, FOE requested that the court “grant plaintiffs such additional and further relief as the court may deem just and proper.” Compl. II8. No purpose would be served in requiring FOE to initiate a new law suit to litigate the scope of the court’s injunction. Thus, the court concludes that under the liberal pleading requirements of Fed.R. Civ.P. 8, FOE’s complaint seeks relief under the NDAA.

Because it claims that the NDAA issue is not before the court, the government chose not to address the issue in detail. Nevertheless, the government does cite a subsequent decision in FOE v. Navy, where the Ninth Circuit found that the Shorelines Board decision constituted issuance of the Shoreline permit, even though state court review was still to come. Friends of the Earth v. Navy, 850 F.2d 599 (1988). There, the Ninth Circuit observed that in the NDAA, “Congress [did not intend] the concept of issuance to become enmeshed with the intricacies of the State of Washington’s judicial review proceedings concerning agency action.” Id., at 601 (attached as exh. 1, Response to FOE’s Motion To Clarify). Based on this aspect of the Ninth Circuit’s holding, the government asks this court to view its setting aside of the 404 permit, as merely suspending the permit, rather than revoking it. Again citing the added cost, the government argues that the equities dictate a suspension, rather than a revocation. In addition, the government contends that once it has cured the NEPA “procedural” violations, the Corps may decide “to reinstate or modify the permit, which would only be possible if the 404 permit is suspended [because] if it is revoked there would have to be an entirely new permit application.”

However, to suspend the permit would not conform with the CWA nor with the spirit and intent of Congress’ prohibition of the use of Homeport funds. EPA guidelines prohibit the Corps from issuing a 404 permit where a practicable alternative exists or where the dredge and fill material will cause or contribute to the significant degradation of United States waters. 40 C.F.R. § 230.11 (a), (c). As the forthcoming memorandum opinion explains, the court has found that the Corps acted unlawfully in finding that no practicable alternative exists and that no significant degra *915 dation will occur. Thus, the Corps’ did not legally issue the Navy’s 404 permit, and by setting it aside, the court held that no permit exists.

Because the court has in essence revoked the permit, as required by the CWA, the Ninth Circuit’s most recent pronouncement concerning the intricacies of judicial review does not apply. Congress specifically decided that no funds be expended in order to “ensure” that all environmental issues be fully addressed. Conf.Rep.H.R. 1748, 1987 NDAA; FOE v. Navy, 841 F.2d at 982. By setting aside and revoking the 404 permit, the court has held that those issues have not been addressed. Thus, to meet Congress’ intent, the court’s injunction must prohibit the expenditure of all Homeport funds until the NEPA and CWA violations have been cured by a thorough, unbiased supplemental investigation and EIS. See FOE v. Navy, 841 F.2d at 927 (to ensure that environmental concerns would be fully addressed prior to construction, “Congress provided only one method to achieve its purposes” — a permanent injunction).

IT IS NOW, THEREFORE, ordered as follows:

The court GRANTS plaintiffs’ motion to clarify, and in conformance with the express mandate of the 1987 and 1988 NDAA, the court permanently enjoins the Navy from obligating or expending any funds for construction of the Everett Homeport until such time as the government complies with NEPA and the Clean Water Act as explained in the court’s forthcoming memorandum opinion.

MEMORANDUM DECISION

Plaintiffs, Friends of the Earth and other environmental organizations (“FOE”), filed a complaint for declaratory and injunctive relief, seeking to halt the United States Navy’s (“Navy”) plans to dredge contaminated sediments in Everett Harbor and deposit them in Port Gardner Bay. FOE contends that the environmental impact statements prepared by the Navy and the Army Corps of Engineers (“Corps”) do not fulfill the statutory mandate of the National Environmental Policy Act and that, in granting the Navy a dredge and fill permit, the Corps violated the Clean Water Act. 1 In a previously issued order, the court concluded that the environmental impact statements did not comply with NEPA and that the Corps had violated the Clean Water Act. Consequently, the court granted summary judgment and enjoined the Navy’s dredging project. 2 This opinion sets out in detail the facts, conclusions, and reasoning underlying the court’s injunction.

FACTUAL AND PROCEDURAL BACKGROUND

A. Factual Overview 3

The Navy plans to build a permanent “Carrier Battle Group Homeport” (“the Homeport”) at Everett, Washington “as part of the Navy’s comprehensive defense strategy.” Friends of the Earth v. U.S. Navy, 841 F.2d 927, 929 (9th Cir.1988) (“FOE v. Navy”). Those charged with making military decisions consider the Homeport to be a critical and integral part of the nation’s maritime strategy. Trost Declaration, at ¶¶ 6-8. To accommodate the Navy vessels, extensive dredging of the *916 harbor’s East Waterway will be required. The Navy intends to dredge approximately 3.4 million cubic yards of sediment from the East Waterway and to dispose of the spoils at a 380-acre site in Port Gardner Bay at depths of 310 to 430 feet. FOE v. Navy, 841 F.2d at 929. Because the area to be dredged has been a repository for industrial wastes, a two to six foot deep layer of thick soup closely resembling “black mayonnaise” covers the floor of the waterway. Thus, “approximately one-third of the dredge spoils are contaminated with heavy metals and organic compounds.” Id. To ensure that all contaminated sediments have been removed, the sediments classified as “contaminated” include this layer of black mayonnaise and an overdredging layer of one to two feet below the layer of black mayonnaise.

The Navy intends to dispose of contaminated and “clean” dredge spoils using a disposal technique called Confined Aquatic Disposal (“CAD”), and refers to the disposal site as “Revised Application Deep CAD” (“RADCAD”). “The CAD disposal system involves in-water disposal of contaminated dredge spoils followed by disposal of clean sediment which, theoretically, will cap and isolate the contaminated material from the marine environment.” FOE v. Navy, 841 F.2d at 929. First, the Navy intends to construct a downslope containment berm at the RADCAD site by dredging “clean” materials using a clam-shell dredge, which will load the dredged materials into barges. The barges will then transport the berm material to the site, and once properly positioned, the barges will open, allowing the sediments to fall to the bottom of Port Gardner Bay. Once the berm has been constructed, contaminated materials will be dredged and deposited at the RADCAD site using the same process. Native sediments lying below the contaminated sediments will then be hydraulically dredged, pumped to the RADCAD site by pipeline, and released under pressure to form a “clean” cap over the contaminated sediments. Disposal will take place in two stages. Phase I involves dredging approximately 836,000 cubic yards of sediment, with 100,000 cubic yards characterized by the Navy as contaminated. Nine months after the completion of Phase I, the Navy intends to begin Phase II, which involves disposal of approximately 828,000 cubic yards of contaminated and 1,636,000 cubic yards of “clean” dredge spoils.

The RADCAD site adjoins an environmentally sensitive area that serves as a breeding ground and provides the primary food supply for a variety of commercially important species of fish and wildlife. United States Fish and Wildlife Service (“FWS”), Impacts of the Proposed Navy Homeporting Project, Everett, Washington (“FWS Report”) 17-34 (January, 1987), Complaint exhibit 3 (“Compl. exh.”). The Ninth Circuit observed that “the CAD method is experimental at [the proposed] depths and the harm to the marine environment which would occur should the contaminated spoils not be contained would be substantial.” FOE v. Navy, 841 F.2d at 929. Because of the environmental sensitivity of the RADCAD site, and a perceived uncertainty of the CAD disposal method, the Navy’s dredging plans generated opposition from FWS and the United States National Marine Fisheries Service (“NMFS”). Id.

B. Prodedural History

1. Federal and State Environmental Requirements

A variety of federal and state environmental laws require the Navy and the Corps to undertake certain actions. Pursuant to the National Environmental Policy Act (“NEPA”), 42 U.S.C. § 4321 et seq., environmental impact statements (“EIS”) were prepared. The Navy prepared a draft and a final EIS (“FEIS”). Before the Navy could discharge dredged or fill material into the navigable waters of the United States, it had to apply to the Corps for various permits including a dredge and fill permit under Section 404 (“404 permit”) of the Clean Water Act (“CWA”). 4 In decid *917 ing whether to issue the various permits, the Corps prepared and issued a draft (“DSEIS”) and a final supplemental EIS (“FSEIS”). With respect to the Navy’s 404 permit application, regulations require the Corps to issue a public notice, 33 C.F.R. 325.3, which triggers a public comment period lasting from 15-30 days. 33 C.F.R. 325.2(d)(2). No dredge and fill permit can issue unless the Corps determines that the proposed project conforms to EPA guidelines set out 40 C.F.R. § 230 Subpart B. When the Corps reaches a decision on the permit application it issues a formal Record of Decision (“ROD”) explaining its rationale and setting forth its findings.

The State of Washington plays a role in this process under both federal and state law. As a prerequisite for obtaining a 404 permit, section 401 of the CWA 5 requires the Navy to obtain a state water quality certification (“WQC”), certifying that Washington’s water quality would not be adversely affected by the activities authorized by the 404 permit. In Washington, the Department of Ecology (“Ecology”) issues a WQC, with a right of appeal to the three-member State Pollution Control Hearings Board (“Pollution Board”). Washington law also required the Navy to obtain a separate shoreline permit from the City of Everett pursuant to the Shoreline Management Act (“SMA”), RCW 90.58.-010-.930. Ecology must approve the shoreline permit, with a right of appeal to the 6-member Shorelines Hearing Board (“Shorelines Board”). The members of the Pollution Board constitute one-half of the six-member Shorelines Board. RCW 90.-58.170.

2. Procedural Chronology

In November of 1984, the Navy published its draft EIS. After holding a public hearing and receiving written comments, the Navy revised its draft and on June 5, 1986, published its FEIS, which contained two volumes of technical appendices and one volume of letters of comment. This report, and the technical appendices on which it relied, focused on the Navy’s preferred CAD site known as Deep Delta, with the majority of the site in 250 feet of water. The Corps determined that a supplemental EIS was required, but just prior to publication of the Corps’ draft supplemental EIS (“DSEIS”), the Navy announced that, in response to concerns regarding dungeness crab populations at the original site, it had moved the Deep Delta site downslope to a new site in water 220-320 feet deep. The Corps published its DSEIS in July of 1986.

In commenting on the draft DSEIS, the FWS criticized the Corps’ analysis on a variety of topics and concluded that the Navy’s CAD proposal “would pose a significant risk to fish and wildlife resources.” FSEIS, XIII-62 (letter dated Sept. 5, 1986). Likewise, the NMFS was very critical, concluding that “open water disposal of contaminated sediments from the East Waterway should not be performed in Port Gardner Bay.” FSEIS, XIII-47 (Letter dated Sept. 4, 1986). Both the FWS and the NMFS believed that upland disposal alternatives, located at Smith Island, offered the best opportunity for limiting potential environmental impacts and required more detailed study. Id.

While the Corps continued work on its EIS, the Navy submitted its third revised application for a “404 permit” on September 29, 1986, identifying yet another location further downslope, the present RAD-CAD site, as its preferred site. Record of Decision, at 4 (compl. ex. 1). On October 7, the Corps issued its public notice concerning the Navy’s third revised application, with a 30 day public comment period. The public comment period on the 404 application ended on the same day as the Corps published its FSEIS, November 6, 1986. The FSEIS text consisted of two volumes, including responses to letters of comment, and a separate volume devoted to technical appendices. 6 The public comment period on the Corps’ FSEIS ended on December 6, 1986, and the Corps itself did not, at any *918 time thereafter, provide further public comment on any aspect of its decision.

In January, 1987, Ecology held its only public hearing on its draft WQC. Thornton Decl. attachment, at 2. The draft WQC required that contaminated sediments greater than 3 centimeters (“cm”) thick be covered by cap material, Administrative Record (“AR”): 8-304, and contained an Appendix A entitled Construction, Operation, and Monitoring Requirements. AR: 8-307. On March 2, 1987, Ecology issued the WQC, but with specific additional conditions. The WQC requires substantial changes in project design, including making the smaller Phase I of the project serve as a pass/fail test for Phase II, and requiring a specific monitoring plan to assure that Phase I succeeded. If the Navy fails to demonstrate that Phase I met certain pass/fail criteria, the RADCAD site cannot be used for Phase II. The WQC required a 1 meter (“m”) cap, but only for contaminated sediments greater than 3 cm thick, AR: 13-341, contained Appendix A, which underwent significant changes from the draft, AR: 13-348, and contained a new Appendix B, setting forth specifics for baseline surveying and a monitoring program to evaluate the environmental effects of Phase I disposal, AR: 13-379. Although the WQC required the Navy to submit a draft monitoring plan incorporating these requirements, it did not set out specifics for monitoring Phase II or the long-term. AR: 13-380. On March 31, 1987, FOE and others appealed to the Pollution Board, seeking to challenge Ecology’s decision to issue the WQC.

The Corps did not immediately adopt the WQC requirements and issue the 404 permit because the FWS and NMFS continued to object to the project. In its report issued in January, 1987, the FWS characterized the CAD proposal as experimental and continued to oppose the project even with the WQC draft requirements. See Compl. exh. 3. In April, the NMFS again asserted its opposition to CAD, contending that because the final WQC requirements were not sufficient, CAD would have “unacceptable adverse impacts on aquatic and fishery resources.” NMFS Letter of April 3, 1987 (Compl. exh. 4). However, after extensive negotiations between the Corps’ North Pacific Division and the resource agencies, held pursuant to 33 U.S.C. § 1344(q), a memorandum agreement was reached on May 21, 1987 (“Memorandum Agreement”). The Agreement contained mutually acceptable specific conditions to be included in the 404 permit in addition to the WQC requirements. NMFS Letter of May 21, 1987 (Compl. exh. 5); AR:12-2 (letter of Sept. 18, 1987). Despite reaching this compromise, the resource agencies continued to recommend upland disposal, and believed that the results obtained from the studies required by the Memorandum Agreement “may well lead us to upland disposal.” Compl. exh. 5.

The City of Everett issued a shoreline permit on June 10, 1987, and Ecology gave its approval on July 8, 1987. FOE and others appealed the issuance of the shoreline permit to the Shorelines Board on July 29, 1987. Because the three members of the Pollution Board also sit on the six-member Shorelines Board, and because the shoreline permit appeal and the WQC appeal concentrated on water quality considerations, the Shorelines Board and the Pollution Board heard the appeals concurrently. During this same time period, the Navy issued its draft monitoring plan, which generated strong criticism from the NMFS, see NMFS Letter of August 28, 1987 (Compl. exh. 10), and by the Corps itself. See Corps Letter of August 27, 1987 (Compl. exh. 11).

On September 1, 1987, while the state administrative appeals were still pending, the Corps’ Chief of Engineers notified the NMFS and FWS of his rejection of a majority of the special conditions required by the Memorandum Agreement as either inappropriate or unnecessary. Letter from E.J. Hatch, Chief of Corps of Engineers, September 1, 1987 (Compl. exh. 6); NMFS Letter of October 5, 1987 (Compl. exh. 7). Because the 404 permit did not include a majority of the conditions contained in the Memorandum Agreement, the FWS and NMFS had “no recourse but to revert to [their] original recommendation for upland *919 disposal of contaminated material” in part because “a number of monitoring actions [were necessary] to determine the effectiveness of untried deep water disposal process.” Letter of Sept. 18, 1987 (AR 12:2). The FWS and NMFS recommended denial of the 404 permit because “the project, as proposed and conditioned, poses a risk of significant adverse effects to fish, shellfish, and the aquatic ecosystem from the discharge of pollutants, and may contribute to the significant degradation of Port Gardner Bay.” Letter of Sept. 18, 1987 (AR 12:3) (emphasis added). Although FWS and NMFS could have taken their objections to the Washington D.C. level of the Corps, they chose not to do so because they thought it futile. Id. at 3.

The Corps issued the 404 permit and its accompanying ROD on September 24, 1987. See Complaint exh. 1. The 404 permit incorporated the WQC conditions including the monitoring plan specifics set out in Appendices A and B of the WQC. However, the Navy’s final monitoring plan was not released until November of 1987, and not approved by the Corps until April of 1988.

After issuance of the 404 permit, FOE sought an injunction to prevent commencement of the dredging project before state administrative bodies had an opportunity to rule on its appeals. On appeal, the Ninth Circuit held that the National Defense Authorization Act required Homeport expenditures to be enjoined until the shoreline permit had been approved by the Shorelines Board. FOE v. Navy, 841 F.2d at 936.

On May 17, 1988, the Pollution Board and the Shorelines Board issued their decisions. By a 2-1 vote, the Pollution Board affirmed the issuance of the WQC. FOE v. Department of Ecology, PCHB Nos. 87-63 and 87-64 (Findings of Fact Conclusions of Law And Order). (“PCHB’) (attached as ex. 1, Opp.Mem.) The Shorelines Board split 3-3, with three members affirming the WQC as issued, adopting the Pollution Board’s findings and conclusions. FOE v. Navy, SHB Nos. 87-31 and 87-33 (May 17, 1988 (Final Findings of Fact Conclusions of Law And Order) (“SHB, Dufford et. al ”) (attached as exhibit 2, Opp.Mem.). The three other Shoreline Board Members issued a competing set of findings and conclusions affirming a permit but placing additional conditions and requirements. Id. (“SHB, Bendor et al”). Under Washington law, a split vote results in affirmance of the WQC as issued by Ecology. 7 FOE then petitioned the Ninth Circuit not to lift its injunction until a state court acted. The Ninth Circuit rejected FOE’s request. Friends of the Earth v. Navy, 850 F.2d 599, 601 (Ninth Circuit June 30, 1988). Most recently, a King County Superior Court refused to lift a stay of the Navy’s permit until such time as the Navy complied with certain additional requirements set out by three members of the Shorelines Board in SHB, Bendor et. al.

C. The Instant Cause of Action

While awaiting resolution of the state permit issues, FOE filed a complaint in this court, seeking declaratory and injunctive relief against EPA, the Corps, and the Navy on issues of Federal law distinct from those considered by the Ninth Circuit. FOE seeks a permanent injunction alleging (1) that the EISs prepared by the Navy and the Corps violated the environmental disclosure and informed decisionmaking requirements of NEPA; and (2) the government’s flawed analysis of the environmental consequences of, and alternatives to, the Navy plan violated section 404 of the CWA, EPA’s implementing regulations, and the Corps’ own public notice requirements. The government responds in depth to each of FOE’s contentions. In addition, the government points out that the Chief of Naval Operations has stressed that expeditious completion of the project is vital to our national security. Trost Decl. II10. As a threshold matter, the government argues that decisions by Washington State administrative bodies should preclude FOE from contesting various factual issues.

*920 II

EFFECT OF STATE ADMINISTRATIVE DECISIONS

The government contends that the state hearing boards performed an exhaustive de novo review of many environmental issues now before the court and rejected FOE’s position on a variety of factual issues. Consequently, the government argues that the court should either (1) grant these decisions preclusive effect because of the doctrines of issue preclusion (collateral estoppel) and comity; or (2) accord the state boards’ findings great deference and dignity. The court finds the government’s arguments unpersuasive. To determine the preclusive effect of state judgments in federal court, the court must apply state law. Fernhoff v. Tahoe Regional Planning Agency, 803 F.2d 979, 986 n. 8 (9th Cir.1986). Washington accords preclusive effect to agency determinations of factual issues if certain conditions are met. Shoemaker v. Bremerton, 109 Wash.2d 504, 507-08, 745 P.2d 858 (1987). In addition, the United States Supreme Court has held that collateral estoppel should not be invoked when doing so denies a party a “full and fair opportunity” to litigate the issue in question. Allen v. McCurry, 449 U.S. 90, 95, 101 S.Ct. 411, 415, 66 L.Ed.2d 308 (1980).

The court finds it inappropriate to give preclusive effect to any factual findings entered by the state boards. The fact that the six-member Shorelines Board issued an evenly split (3-3) decision makes untenable the government’s argument that the Shorelines Board rejected FOE’s factual contentions. In fact, three board members would not have allowed the Navy’s dredging plan to proceed as presently designed because “without additional shoreline permit criteria [the RADCAD disposal] will likely ... cause chronic long-term toxic effects to the aquatic life of Puget Sound.” SHB, Bendor et. al. at 46. In any event, an even split of the board does not constitute a binding decision, and merely leaves the permit standing as issued. Hayes v. Yount, 87 Wash.2d 280, 295-96, 552 P.2d 1038 (1976). Consequently, collateral estoppel principles do not apply. Shoemaker v. Bremerton, 109 Wash.2d at 507-08, 745 P.2d 858; see also, Peterson v. Department of Ecology, 92 Wash.2d 306, 313, 596 P.2d 285 (1979) (collateral estoppel inapplicable where administrative board’s decision is “ambiguous and inconsistent”).

Even though the Pollution Board did unambiguously uphold the WQC by a 2-1 vote, the Board merely resolved state legal issues. It did not, and could not, address the Corps’ responsibilities under section 404 of the CWA or the sufficiency of the federal EISs under NEPA. Collateral estoppel does not apply where the second proceeding involves different legal standards, even though the facts are identical. Clark v. Watchie, 513 F.2d 994, 999 (9th Cir.), cert. denied, 423 U.S. 841, 96 S.Ct. 72, 46 L.Ed.2d 60 (1975); Peterson v. Clark Leasing Corp., 451 F.2d 1291, 1292 (9th Cir.1971) (per curiam). A state administrative decision does not have preclusive effect on federal court review of a 404 permit because the state body “could not relieve the Corps of its obligation under NEPA or bind it in its determinations.” Sierra Club v. Alexander, 484 F.Supp. 455, 464 (N.D.N.Y.), aff'd mem., 633 F.2d 206 (2d Cir.1980). Under these authorities collateral estoppel would not be appropriate in this case.

Moreover, according preclusive effect to either board’s decision would be inappropriate because the record before the court contains a variety of evidence not reviewed at the state level. 8 To grant preclusive *921 effect under these circumstances would contravene public policy and work an injustice on FOE, circumstances that prevent the application of collateral estoppel. Shoemaker v. Bremerton, supra; see also Allen v. McCurry, 449 U.S. at 95, 101 S.Ct. at 415.

Nor does the court believe that it must accord deference to the Pollution Board’s findings. The government bases its deference argument on the conclusive effect given to the WQC on water quality issues. See 33 U.S.C. § 1371(a); see also Friends of the Earth v. Hintz, 800 F.2d 822, 834 (9th Cir.1986) (under Corps regulations, 33 C.F.R. § 320.4(d), a WQC certification is conclusive with respect to water quality considerations) {“FOE v. Hintz ”). However, 33 U.S.C. § 1371(a) expressly refers to activities under the Rivers and Harbor Act, and makes no reference to the CWA. Even though the Ninth Circuit referred to the conclusive effect of a state certification under corps regulations with respect to “water quality,” the court did not address whether this applies to the various other EPA criteria governing the issuance of a 404 permit, see 40 C.F.R. § 230.10(a)-(d). Moreover, legislative history reveals that by requiring a state WQC, Congress merely intended to assure that a 404 permit would reflect state water quality standards, and that section 401 was not intended in any way to supplant requirements for obtaining a 404 permit. Monongahela Power Co. v. Marsh, 809 F.2d 41, 53 n. 114 (D.C. Cir.1987). Finally, the Ninth Circuit in FOE v. Hintz held that the appellants were foreclosed from raising the water quality issue with respect to the 404 permit because, unlike FOE, the Hintz appellants “did not challenge [Ecology’s] certification, or the preclusive effect of this regulation, in any proceeding below.” 800 F.2d at 834. 9

For the above reasons, the court declines to accord preclusive effect or deference to the various state board decisions. Refusing to view either set of findings and conclusions with deference does not mean, however, that the court will not review the state board decisions to determine whether their respective analyses help explain the federal administrative record now before the court.

Ill

NEPA CHALLENGE

A. Standard of Review

Review of agency determinations under both NEPA and the CWA is generally limited to the administrative record. FOE v. Hintz, 800 F.2d at 828-29; Animal Defense Council v. Hodel, 840 F.2d 1432, 1436 (9th Cir.1988). Under certain circumstances, the court may consider evidence outside the record. Animal Defense Council, 840 F.2d at 1436. Circumstances relevant here include (1) “when necessary to explain the agency’s action,” Id.; (2) when “necessary to explain technical terms or complex subject matter involved in the agency action,” Id.; and (3) when necessary “to ascertain whether the agency considered all relevant factors or fully explicated its course of conduct or grounds of decision,” FOE v. Hintz, 800 F.2d at 829. FOE and the government offer a variety of extra-record evidence, including the decisions of the state boards and declarations of various scientists and engineers. In most instances, the extra-record evidence explains the data and factors on which the Navy and the Corps relied, and thus can be relied upon by the court.

Judicial review of an agency’s preparation of an EIS is governed by the Administrative Procedure Act (“APA”), 5 U.S.C. § 706(2)(D) (1982), under which the court may set aside agency action undertaken “without observance of procedure required by law.” Northwest Coalition for Alternatives to Pesticides v. Lyng, 844 F.2d 588, 590-91 (9th Cir.1988). Because *922 “NEPA is essentially a procedural statute”, the court can not “substitute its judgment for that of the agency concerning the wisdom or prudence of a proposed action.” Id. Therefore, the adequacy of an EIS depends upon the rule of reason; “a reviewing court may not ‘fly speck’ an EIS and hold it insufficient on the basis of inconsequential, technical deficiencies.” Methow Valley Citizens Council v. Regional Forester, 833 F.2d 810, 815 (9th Cir.1987) (citations omitted).

Thus, the issue before the court is not whether CAD will work or whether the court would choose CAD as a disposal method. Rather, the court must determine whether the EISs satisfy their primary purposes of (1) providing decisionmakers with “an environmental disclosure sufficiently detailed to aid in the substantive decision whether to proceed with the project in light of the environmental consequences,” and (2) providing “the public with information and an opportunity to participate in gathering information.” Methow Valley, 833 F.2d at 814. FOE argues that the Corps and Navy failed to meet their NEPA obligations with respect to three general areas: (1) technical/scientific issues; (2) treatment of a proposed upland alternative; and (3) reasonable mitigation measures.

B. Technical/Scientific Issues: Scientific Uncertainty and Controversy

FOE contends that the Navy and the Corps violated NEPA by failing to disclose and address uncertainty and scientific controversy as to CAD’s technological feasibility and the significant risk of environmental harm it poses. In response, the government argues that all environmental issues were aired and the project as presently constituted “poses no environmental risks.” Opp. Mem. 12. The court will address the technical/scientific issues separately, keeping in mind that NEPA “does not require [a court] to decide whether an EIS is based on the best scientific methodology available, nor does NEPA require [a court] to resolve disagreements among various scientists as to methodology.” Oregon Environmental Council v. Kunzman, 817 F.2d 484, 4966 (9th Cir.1987). Moreover, when the Corps issues factual findings involving highly technical areas of expertise, the court must accord them a high degree of deference. Sierra Club v. Froehlke, 816 F.2d 205, 215 (5th Cir.1987). A federal court is not in the business of resolving scientific disagreements between plaintiffs’ experts and the Corps’ experts. Story v. Marsh, 732 F.2d 1375, 1381 (8th Cir.1984). Nevertheless, the court must ensure that the Corps and the Navy adequately considered and disclosed the environmental impact of the dredging project. Baltimore Gas & Electric Co. v. NRDC, 462 U.S. 87, 97-98, 103 S.Ct. 2246, 2252, 76 L.Ed.2d 437 (1983).

1. Technological Feasibility: How “Certain” is “Certain”?

FOE contends that the Corps’ treatment of CAD’s technological feasibility violated NEPA in two ways: (1) the Corps failed to disclose in the text of the FSEIS the experimental nature and uncertainty surrounding the RADCAD project; and (2) relying on its conviction that the project would be successful, the Corps failed to assess the environmental impact of a technological failure. In response, the Corps contends that the project’s careful design, including extensive monitoring and conservative assumptions, negates the importance of any alleged uncertainty. After careful review of the record, the court concludes that the Navy’s and the Corps’ EISs violated “the touchstone” of NEPA obligations, failing to disclose sufficient information to ensure that “informed decision-making and informed public participation” will take place. State of Cal. v. Block, 690 F.2d 753, 767 (9th Cir.1982).

The text of the various EISs does not reveal the forthright admission contained in the ROD issued almost a year later: the CAD technology has “potential for major environmental impacts ” that can be avoided “if the RADCAD operation works according to design.” AR: 13-256 (emphasis added). Despite this admission in the ROD, careful scrutiny of the entire administrative record before the court reveals *923 that the Corps did not identify and discuss the specifics of the “major environmental impact” that would result if CAD does not perform as predicted. 10 No disclosure took place because the Corps displayed an unwavering certainty in CAD’s ability to perform as predicted and total reliance on the WQC monitoring plan to avoid any environmental consequences in the unlikely event that the technology fails to perform as expected. However, the court finds that, in contrast to the certainty of RADCAD’s success displayed in the text of the EISs, the technical appendices and the real world experience with CAD demonstrate that the disposal plan is experimental, subject to a significant degree of uncertainty, and presents a significant risk of failure.

The Corps’ EIS characterizes the RAD-CAD disposal project as merely involving an “extension” of existing technology, FSEIS, III — 62, maintaining that “the only identifiable reason for cap failure is insufficient volume of cap materials.” Id. at II-31 (emphasis added). In its briefing to the court and in its oral argument, the Corps continues to assert unconditionally that the capping technology will perform as predicted. See, e.g., Opp. Mem. Sec. 11(B)(2). In support of its certainty of success, the Corps asserts that placement technology is based upon modelling and state of the art techniques which will be monitored to assure proper cap placement. Yet, close scrutiny reveals that the Corps’ own data and conclusions do not support its dogmatic belief in the certainty of RADCAD’s success.

The Corps bases its predictions of sediment behavior on research and analysis gathered from three field sources: (1) successful examples of CAD technology; (2) a number of open water, uncapped disposal sites at various depths, FSEIS, App. B., at 12-17; and (3) a Yale University study that found the same processes govern material disposal behavior at depths ranging from 60 to 220 feet, see FSEIS, App. B., at 27. In addition to field results, the Corps also conducted a numerical modeling study using a model originally developed by EPA, known as the DIFID model. Based on its modelling and analysis of field data, the Corps’ concluded that the laws of physics governing the descent of disposal materials are consistent at any depth, and therefore “the behavior [of disposed sediments] can be appropriately estimated [at any depth].” Palermo Decl. 1163

Although the various field studies and modelling support the Corps’ contention that RADCAD is technologically feasible in the abstract, the acknowledged limitations of this data demonstrate that CAD technology is at best experimental and uncertain at RADCAD depths. For example, the Corps acknowledges that DIFID modelling was not conducted for RADCAD conditions (310-430 feet deep). Rather, modelling concerned conditions existing at the Deep Delta, 265-foot-deep site, and the Corps merely made mathematical adjustments to take account of the “slightly deeper” water at RADCAD. Palermo Decl., at 20. The Corps further acknowledges that DIFID modelling has severe limitations, being incapable of predicting sediment behavior for more than a single bargeload dump. In fact, no sophisticated computer model exists to predict mounding (the size and shape of the disposal area) that would result from a large volume of material from multiple dumps. FSEIS, App. B. of App. B, at 18. Yet, RADCAD involves barges having a 4,000 cubic yard capacity. SHB, Bendor et. al, at 8. Thus, Phase I of the project involves 25 bargeloads of contaminated sediment, while Phase II involves approximately 200 bargeloads. The absence of any model capable of predicting two bargeload dumps, let alone 25 or 200, inter alia led three members of the Shorelines Board to characterize the CAD disposal method as experimental in “significant ways.” Id. at 29.

Further evidence of CAD’s experimental nature comes in the form of the Corps’ *924 “successful” use of the technology. Barge-dump capping has succeeded only in 70 foot-deep water, while the Navy’s proposal involves CAD at depths from four to six times greater. In fact, CAD has not been attempted in this country at depths greater than 100 feet. Compl. exh. 4, attachment, at 1. In a 1984 test of CAD in the West Waterway of Elliot Bay, the Corps attempted to place a single barge-load (1,1100 cubic yards) of contaminated sediments in a depression in 70 feet of water. While the Corps terms this test successful, its own monitoring established that a sizeable amount of the barge-load surged out of the depression. Id.; SHB, Bendor et. al, at 30.

Furthermore, attempts at mound formation (without capping) in depths approaching those of the RADCAD site proved unsuccessful, apparently because of an inability to accurately place the material at the site. FWS Report, at 1. The Corps acknowledges this “failure”, FSEIS, III-38, which occurred at the Foul Area Disposal (“FAD”) site off New England in water ranging from 160-300 feet deep. Palermo Decl., at 31; Germano Dec!., at 12-17. However, the Corps maintains that the actual cause of the problem was a failure of the barges to dump at the pin-point sites, Germano Deck, at 14. The Corps’ assures the court that due to the WQC’s tight control over dumping operations there “should [be] no problem similar to that observed at the Foul Area site.” Palermo Deck, at 32. Nevertheless, the fact is that FAD, the only attempt of mound formation near RADCAD depths, failed.

Finally, RADCAD’s experimental nature is highlighted by the fact that no field data exists for hydraulic cap placement because it has never been attempted. The Corps’ Waterways Experiment Station (“WES”) summarized RADCAD technological considerations, and acknowledged the existence of “gaps or shortfalls in existing technology” with respect to hydraulic capping and the lack of models to determine the “accumulative behavior of multiple dumps.” Palermo, et. al, Evaluation of Dredged Material Disposal Alternatives for US Navy Homeport at Everett, Washington, Tech. Rep. EL-88, U.S. Army Engineer Waterways Experiment Station, at 130 (February, 1988) (“WES Evaluation ”). Even though the WES Evaluation expressly notes that conservative assumptions were used where technological gaps exist, the fact remains that these technological gaps and uncertainties clearly exist, and the EISs create a false sense of security by painting a picture of absolute certainty-

While technically feasible, the RADCAD proposal remains experimental in the eyes of a wide variety of knowledgeable observers. See SHB, Bendor et. al, at 29-31; FWS Report; NMFS Letter of April 3, 1987 (compl. exh. 4). Moreover, as the Corps acknowledges, it is impossible to be certain of RADCAD’s success because “there is no way to predict with absolute certainty what will occur during construction.” FSEIS, App.B., at 15. Based on the conclusions of the FWS and the NMFS, the Ninth Circuit found the CAD method to be “experimental” at RADCAD depths and recognized that “substantial” harm would occur to the marine environment should the contaminated spoils not be contained. FOE v. Navy, 841 F.2d at 929.

The body of the EISs did not reflect any of the concerns raised by the FWS and NMFS. The resource agencies’ views were only presented in the comments and response section, a placement the government contends is adequate. The court disagrees. The concerns and views of these key agencies should have been contained in the body of the EISs. Opposing views must be reflected at an “appropriate point.” 40 C.F.R. § 1502.9(b). Whether the comments and response section constitutes an “appropriate point” depends, in part, upon whether the body of the EIS discusses the environmental problems raised by opposing views and, in the case of federal agencies, whether the opposition is merely a substantive comment, rather than opposition to the project itself. Oregon Natural Resources Council v. Marsh, 832 F.2d 1489, 1499, n. 11 (9th Cir.1987) (“ONRC v. Marsh ”). The court concludes that, for two distinct reasons, the resource *925 agencies’ views should have been included in the body of the FEIS and the FSEIS. First, the FWS and NMFS clearly and repeatedly stated strong opposition to the project itself, rather than merely making substantive comments. Second, the body of the FEIS and the FSEIS never discusses the environmental problems raised by any responsible party, let alone the views of these key Federal agencies.

In response to the position that the FSEIS should have clearly acknowledged the myriad of uncertainties surrounding the CAD technology and identified the major environmental impact a failure would cause, the Corps maintains that the success of the project is guaranteed by the monitoring requirements set down in Washington’s WQC, and subsequently adopted by the Corps in the ROD issuing the 404 permit. Recently, the Ninth Circuit held that even though a Corps EIS was insufficient under NEPA, the ROD represents the Corps’ final decision, and where the ROD adopted suggestions made by EPA that eliminated the areas of environmental concern, the Corps cured its original failure to obtain information necessary to make a reasoned decision. Half Moon Bay Fishermans’ Marketing Ass’n v. Carlucci, 847 F.2d 1389, 1395-97 (9th Cir.1988).

Here, although the Corps issued a detailed ROD in support of its decision to issue a 404 permit, the court concludes that the ROD did not eliminate the areas of environmental concern, and therefore Half Moon Bay does not apply. 11 Although the Corps contends that the WQC requirements adopted in the ROD ensure RADCAD’s success, the monitoring plan does not alter CAD’s experimental nature or remove the possibility of a technological failure. For example, the rejected Memorandum Agreement with the NMFS and the FWS called for a performance criterion requiring 90 per cent of the berm mound to be confined in the area identified by the Navy’s drawings. NMFS Letter of May 21, 1987, at 2 (Compl. exh. 5). Yet, the Corps’ Chief of Engineers rejected this criterion, relying instead on the WQC requirements. Compl. exhs. 6 and 7, at 2-3. The Corps’ total reliance on CAD’s accuracy in placing the sediments in their intended resting place is inconsistent with its refusal to set a 90 per cent placement performance criterion, especially given the Corps’ assurances that no more than 4.5 percent of the material will be released during dredging and disposal. Letter of NMFS, Compl. exh. 7, at 3. If, as the Chief Engineer has stated, it would be difficult to measure whether 90 per cent of the mound volume was accurately placed, id., the court questions how post-disposal monitoring could possible determine the success of the placement technology and insure against environmental injury. Moreover, the WQC performance criteria, relied upon so heavily by the Corps, will declare Phase I placement a success even if the contaminated material spread to 250 per cent of the area specified in the FSEIS. Compl. exh. 7, at 6.

The Corps also contends that placement of the berm material is a confirmatory test of placement technology. In fact, the berm stage is not a pass/fail test for Phase I, which would proceed “unless no discernible berm whatsoever is detected.” SHB, Ben-dor et al., at 9-10.

The court acknowledges that having the much smaller Phase I serve as a pass/fail test for Phase II would lessen the harm to the environment should Phase I fail. The record, however, continues to lack any analysis of the environmental consequences of a failure of Phase I or Phase II.

In summary, the court concludes that the Corps’ and the Navy’s EISs (1) failed to acknowledge the degree of uncertainty concerning the CAD technology and its use at RADCAD depths; (2) failed to appropriate *926 ly reveal the views of the federal resource agencies on the subject, and (3) failed to identify the “major” environmental consequences of a technology failure. NEPA requires an EIS to expose scientific uncertainty concerning safety and environmental risk of a proposed action. Southern Oregon Citizens Against Toxic Sprays, Inc. v. Clark, 720 F.2d 1475, 1479 (9th Cir.1983), cert. denied, 469 U.S. 1028, 105 S.Ct. 446, 83 L.Ed.2d 372 (1984) (“SOCATS ”). Moreover, an EIS “must be particularly thorough when the environmental consequences of federal action are great.” Warm Springs Dam Task Force v. Gribble, 621 F.2d 1017, 1026 (9th Cir.1980). Although an EIS “need not discuss remote and highly speculative consequences,” id., the court concludes, along with the Ninth Circuit, FWS, NMFS, and three members of the Shorelines Board, that the RADCAD project is experimental and fraught with uncertainties. In such a case, the “major” environmental consequences that would result from a failure cannot be said to be “remote and highly speculative.” Because the EIS text failed to disclose and discuss crucial information concerning technological uncertainty and what major environmental impact would occur if the RADCAD technology failed, the Corps and the Navy could not possibly have satisfied NEPA’s requirement of informed decisionmaking and informed public participation. Half Moon Bay, 847 F.2d at 1394-95; Methow Valley, 833 F.2d at 818; see also Baltimore Gas & Electric Co., 462 U.S. at 97-98, 103 S.Ct. at 2252 (NEPA requires agency to take hard look at environmental consequences and disclose the environmental impact of its action to the public); ONRC v. Marsh, 832 F.2d at 1497 (when information essential to a reasoned choice is lacking and the cost of obtaining it is not exorbitant, the agency must obtain it and include it in the EIS).

2. Sediment Characterization: How “Clean” is “Clean”?

As noted above, the CAD proposal involves using “clean” native sediments to construct the containment berm and to cap contaminated sediments. All told some 2.3 million cubic yards of “clean” native sediments will be dumped or hydraulically pumped into Port Gardner Bay. Because this massive volume of “clean” sediments will remain unconfined, it is “critical” that the berm and capping material “be, in fact Clean, and not have chemical concentrations likely to cause acute or chronic long-term toxicity to marine life.” SHB, Bendor et al, at 19 (emphasis in original). FOE argues that, in characterizing the sediments as clean, the government violated NEPA in two different ways: (1) it lacked critical information necessary for it to make an informed, reasoned decision; and (2) its own data contradicted its conclusion that the native sediments posed no environmental risk. After careful review and study of the administrative record, the various declarations filed by the parties, 12 and the recently published study commissioned by EPA, PTI & Tetra Tech, Everett Harbor Action Program: Analysis of Toxic Problem Areas (June, 1988) {“Tetra Tech/PTI Report ”), the court finds as follows: (1) the sediment characterization tests, conducted by the Navy and the Corps, at best proved inconclusive and recommended further testing, and may in fact directly contradict the EIS and ROD characterization of the native sediments as “clean”; and (2) the tests themselves were insufficient to assess the environmental risk posed by the “clean” sediments.

*927 To determine whether use of native sediments for cap and berm material posed environmental risks, the Corps and the Navy employed Battelle Northwest Laboratories to conduct a three-phased “sediment characterization” study (“Battelle I, II, and III”). Characterization of native sediment as “clean” involved both chemical and biological tests. Crecelius Decl. ¶ 6.

a. Chemical Analysis of Native Sediment

Both Battelle I and Battelle II conducted chemical analysis. Battelle I chemical analysis focused particularly on specific trace metals, pesticides, polynuclear aromatic hydrocarbons (“PAHs”), and chlorinated hydrocarbons (particularly PCBs), each of which had been identified as present in the East Waterway by previous EPA studies. Crecelius Decl., at 6; Anderson Deck, at 10. Bulk chemical analysis was also conducted with respect to the entire list of EPA priority pollutants, see 40 C.F.R. Part 423. FEIS, Battelle I Final Report, at 4 (attached to App.A. of App.P.). Based on these results, Battelle I concluded that the native material was “relatively” clean, in that the level of the pre-selected contaminants was only “slightly elevated” over Puget Sound background. FEIS App.P, Battelle I, at 14; FEIS, App. B. of App.BB, Battelle II, Introduction, at l.B. Battelle II aimed at determining whether the native sandy material was clean enough to be used to cap the contaminated upper layer. Chemical analysis again was limited to EPA’s list of priority pollutants.

The court finds the chemical tests relied on by the Corps and the Navy to be seriously deficient. With respect to the actual chemical analysis conducted in Battelle I and II, the test results themselves do not support characterizing the native sediments as clean and risk free. For example, Battelle I concluded that its tests could not be used to determine whether the native material would be acceptable for unconfined water disposal. FEIS, App.P., at 11. Further biological testing was required. Id. Moreover, after three weeks of testimony, three members of the Shorelines Board concluded that the Battelle I test results demonstrated that “contamination levels in some areas of the harbor increased, rather than decreased with sediment depth.” SHB, Bendor et al, at 25.

The scope of the chemical tests presents another serious deficiency, perhaps of even greater concern. As Dr. Malins points out, the government limited its chemical analysis to conventional pollutants found on EPA’s priority listing. Malins Deck 1123. In fact, Battelle I did recognize the existence of an “unresolved complex mixture”, and recommended further analysis to identify and quantify the unknown chemicals. FEIS App.P., at 34. Dr. Malins states that, by limiting analysis to EPA priority pollutants, the Navy did not test for chemicals that unquestionably are present in the Everett sediments and identified by the Bat-telle tests as the “unresolved complex mixture.” Malins Deck ¶ 31. In particular, he points to a range of compounds called “polar fraction,” often found in pulp mill effluent, an historically significant source of contamination in the Everett East Waterway. Id. While he admits the difficulty of individually identifying polar compounds using chemical analysis, Dr. Malins contends that recent research shows the toxicity of such compounds and that there are scientifically accepted, feasible methods for assessing the total polar fraction of contaminants in the Everett sediments. Id.

In response, Dr. Crecelius notes that 10 polar compounds appear on the EPA priority listing, and that the Navy specifically looked for phenol and cyanide because of their use in industry near Everett. Creceli-us Deck, at 13-15. Dr. Anderson notes that little research has been done on polar compounds because they are difficult to measure and there is no evidence that they are important toxicologically. Anderson Deck, at 15-16.

The recently published Tetra Tech/PTI Report, commissioned by EPA as part of ongoing studies of Puget Sound contamination, sheds considerable light on this dispute between experts. The Report identified high concentrations of more than a dozen polar chemical compounds in East *928 Waterway sediments, compounds known to be associated with pulp mill effluent. Tetra Tech/PTI Report, at 272; Malins 2d Deck ¶ 6. Biological impacts were observed and related to pulp mill discharges. Tetra Tech/PTI Report, at 241. Dr. Malins also cites to a number of peer-reviewed studies linking these same polar compounds to DNA damage and mutagenic effects. Malins 2d Deck 116 and Ex. C (1) and (2). Moreover, the Navy’s Battelle studies did not attempt to measure or identify a number of the most concentrated polar compounds now known to be present in East Waterway sediments. Tetra Tech/PTI Report, at 141-42.

The government argues that the Report does not constitute significant new information requiring supplementation of the record and does not raise serious questions about the distribution of polar compounds in the “clean” sediments. Suppl.Opp.Mem. at 6. The government seeks to distinguish the new findings on two bases: (1) the result stations showing the highest concentrations of polar compounds will not be dredged in the Navy project, Compare Tetra Tech/PTI Report, Figure 5, at 35 with FSEIS, Figure 2.4, at 11-27; and (2) the draft report only analyzed the top two centimeters of East Waterway sediments, Tetra Tech/PTI Report, at 45. The court concludes, however, that the government misses the significance of the Tetra Tech/PTI findings.

The Tetra Tech/PTI Report did in fact find concentrations of many of the polar fraction chemicals in areas to be dredged. Detectable concentrations of many of these chemicals, concentrations in excess of the highest concentrations in the study’s Puget Sound reference areas, were found in station-testing-sites in areas to be dredged for “clean” berm material and in areas to be dredged for contaminated sediments and underlying capping sediments. 13 As Dr. Malins points out, even though only surface sediments were tested, the results at minimum raise serious questions concerning the vertical distribution of these polar compounds because the pulp and paper industry is one of the oldest sources of pollution in Everett Harbor and the East Waterway. Thus, it is quite possible that contamination from polar compounds would also be found in the deeper native sediments now classified as clean. Malins 2d Deck 1I1Í6-7.

Perhaps most importantly, the government’s arguments do not counter the support that the Tetra Tech/PTI Report provides for Dr. Malins criticisms of the government’s research efforts. Dr. Malins correctly points out the following: (1) the Report results undercut Dr. Crecelius’ and Dr. Anderson’s assertions that previously unresolved and unidentified chemicals can be ignored because they are not likely to be toxic; (2) polar chemicals associated with pulp mill effluent can be analyzed without too much difficulty; and (3) the Report results cast doubt on whether prior chemical analysis focused on a far too narrow range of potentially toxic chemicals, in that EPA, as part of its ongoing study of Puget Sound contamination, commissioned this study of compounds not listed on its priority pollutant listing.

b. Biological Testing of Native Sediments

Biological testing conducted during Bat-telle II and III suffered from similar deficiencies. The Battelle studies conducted two types of biological tests: bioaccumulation and bioassay. Bioaccumulation involves exposing test organisms to the sediments and determining the concentrations *929 of pre-selected chemical compounds in their tissue as opposed to the chemical found in tissue of organisms not exposed. Anderson Decl., at 6. Thus, the bioaccu-mulation tests are inherently limited to the chemicals that are pre-selected, which of course did not include any of the polar compounds discussed above. Bioassay actually tests toxicity by exposing an organism to a contaminant for a period of time and observing the results as compared to the same organism exposed to “clean” water or sediment or food. Id. at 7-8.

Results of the Battelle biological studies do not support characterizing the native sediments as “clean” and risk free. Bat-telle II exposed recognized bioaccumulator indicator species, clams and mussels, to sediments from Sequim Bay (control group) and to six composite (mixed) samples collected from the native sediments in the East Waterway. FEIS, App.B of App.BB, at 18-19. Battelle II test results showed a high concentration of chemicals, especially PAH’s. FEIS, App.B of App.BB Table 10, at 20. The government acknowledges this but states that “overall the native material produced far less contamination than the top [heavily contaminated] portions.” Opp. Mem.22. Nevertheless, the levels of contamination in the composite samples far exceeded that in the Sequim Bay control group. Specifically, the “[bioaccumulation] results showed significant chemical accumulation levels from the Everett bottom ‘clean’ sediments, with a PAH level in one instance 16 times the level found in the control group.” SHB, Bendor et al, at 27. In testimony before the state boards, the government’s expert, Dr. Anderson, characterized the test results as demonstrating an “important” and “meaningful” accumulation of PAHs in accumulator test organisms exposed to three of the six composites. True Declaration, exh. 1, at 66.

Battelle II also conducted a standard “acute toxicity amphipod bioassay test”, originally using an amphipod (a type of sand flea) purchased from a supplier in Oregon. The amphipods were exposed to the composite samples and the Sequim Bay control sample for 20 days, after which the surviving amphipods were counted to determine the level of mortality. Thus, this was a test of the acute effects of exposure to the “clean” sediment; no test was conducted to determine the chronic effects of longer exposure (i.e. effects on reproduction, lifespan, etc.). Here again the results do not support the Corps’ unqualified characterization of the berm and cap material as clean and risk free. First, the amphipods originally exposed to the control group had a mortality rate as high as 63 per cent. A second control group of amphipods gathered from a Washington location was then tested, with seven of nine test groups having a mortality rate ranging from 25-45 per cent. Mortality in the composite native samples reached 60% on several occasions and exceeded the control group in most cases. FEIS, App.B of App.BB Table 12, at 24. However, because the control group often did so poorly, and pass/fail criteria was based on the difference in performance, the test group was determined to have passed. Id.

Battelle II concluded that further biological tests were necessary. FEIS, App.B of App.BB., at 32. The Navy’s EIS considered the Battelle II tests as preliminary. FEIS, IV-49. Consequently, Battelle III began in May of 1986, but was never mentioned in the EIS text. Battelle III ran chemical analysis as well as bioaccumulation tests and bioassay tests on amphipods, oyster larvae, geoducks, as well as a Microtox bioassay. DSEIS, App.C to App.B., at 17, 19, 21, 22-24; SHB, Bendor et. al, at 19, 29. Generally, chemical analysis showed the “clean” sediments to have a measure of contamination. However, the Corps rejected the Battelle III results because it concluded that due to the collection method, the “native” sediments were actually mixed with contaminated top sediments. Opp.Mem. 23 n. 24.

Despite the problem with the Battelle III chemical analysis, no new tests were attempted, ostensibly because the native sediments, even when mixed with contaminated sediments, passed the amphipod bioassay tests. Id. at n. 25; Anderson Deck, at 15. However, based on the Corps’ own technical appendices, the court finds that in reali *930 ty the amphipod test-results provide no support for the government’s position. Because the amphipods hardly reacted to contaminated sediments, Battelle III concluded that “[i]t would appear that the amphipods were not very sensitive to test sediments and other approaches would be helpful in defining the potential effects of the material.” DSEIS, vol. 1, App.C. to App.B, at 17. This conclusion confirmed Dr. Malins’ warning, issued in 1984 while he served as a division director for NMFS, that “amphi-pods are not that sensitive to Everett harbor sediment; therefore the proposed study to use amphipod bioassays alone will probably not provide useful information.” FSEIS, App.P, exh. 1, Letter of Oct. 12, 1984 117. Despite this known lack of sensitivity, no further tests were conducted. The government continues to rely on am-phipod bioassay results from Battelle II and III to characterize the native sediments as sufficiently clean for unconfined disposal; a characterization contradicted by Bat-telle III oyster larvae, geoduck, and Micro-tox bioassays, all of which demonstrated that the “clean” sediments had a measure of toxicity. DSEIS, APP.C to App.B, at 17, 19, 21; SHB, Bendor et. al, at 29.

Just as in the case of chemical analysis, the problems with the biological testing go beyond the fact that the test results do not support the government’s conclusions. Dr. Malins points out that the government severely limited the scope of the tests. For example, the court finds troublesome the government’s absolute reliance on test organisms — clams, mussels, and sand flea amphipods — that bear no demonstrated relationship to the Port Gardner ecosystem. Dr. Anderson responded to these concerns, acknowledging that, although the bioacu-mulator test organism (mussels and clams) did not inhabit the RADCAD site, using native species often leads to ambiguous results and the organisms chosen have been demonstrated to be good bioacumula-tors, widely accepted as good surrogates for the native species. Anderson Deck, at 18. Nevertheless, EPA regulations for evaluating biological effects in this situation state that while biological indicator species may be useful, when toxicity tests are required, sensitive native species should be identified as possible bioassay organisms. 40 C.F.R. § 230.61(e)(2). Dr. Malins assures the court that he has used this approach, that it is not difficult, and does not lead to ambiguous results.

Perhaps of even greater concern, is the government’s absolute reliance on biological tests for acute lethality when chronic biological harm from long-term, low-level exposure to sediment contamination is the most probable biological risk according to Dr. Malins. As to testing for chronic effects, Dr. Anderson states that he is “unaware of any standard test for chronic effects of contaminants in sediment.” Anderson Deck, at 17-18. Such tests would by their very nature require months or years to run, and were “beyond the scope of the Everett study.” Id. Dr. Ma-lins claims that such tests are readily available, that he personally has been involved in such tests, and that they could produce useful results in six to twelve months. Ma-lins 2d Deck, at 11-12. In any event, under NEPA, the inability to conduct tests does not allow the government to ignore the possible chronic effects.

c. Conclusion: Not “Clean" Enough

At each stage of the environmental review process, the Corps and Navy relied on Battelle chemical and biological test-results to maintain that the berm and capping sediments pose no threat to the marine environment. Opp.Mem. 24; FSEIS, III — 12 (“[c]hemical and biological analyses have shown that these sediments meet requirements for unconfined water disposal in Port Gardner”); FEIS, IV-49 (concludes that native sediments are not toxic to marine organisms); ROD, at 1 (adopting the EIS findings, and viewing the native sediments as “clean”). The court concludes that the Corps’ treatment of the risk posed by the berm and cap sediments violates NEPA’s requirement of informed decision-making and informed public participation in several ways.

First, the actual Battelle tests results do not support the Corps’ conclusions that the native sediments are sufficiently clean to use as capping material and do not pose a *931 significant risk to the biota. At best, both the chemical and biological tests provide inconclusive results and recommend further testing. When Battelle III results demonstrated that the native sediments were in fact contaminated, the Corps explained away the problem, but never conducted further tests. At worst, the results of the chemical and biological tests suggest that the native sediments may in fact be contaminated. When the technical appendices do not support or contradict the conclusions set forth in the text, the court must invalidate the EISs in question. See, e.g., Northwest Indian Cemetery Protective Ass’n v. Peterson, 565 F.Supp. 586 (N.D.Cal.1983), aff'd, 795 F.2d 688 (9th Cir.1986), reversed on other grounds, — U.S. -, 108 S.Ct. 1319, 99 L.Ed.2d 534 (1988). Moreover, the FSEIS text failed to disclose and address scientific uncertainty concerning the validity and accuracy of the Bat-telle test results and the safety and environmental risk posed by the native sediments. These omissions clearly violate NEPA. SOCATS, 720 F.2d at 1479.

Second, the body of the EISs fails to reflect the opposing views of FWS, NMFS, and other responsible scientific authorities with respect to the potential toxicity of the native sediments. Here again, the court concludes that addressing these views in the comments and response section does not meet the Corps’ NEPA obligations. Because the body of the EIS never mentions, let alone discusses the environmental problems, and because the FWS and NMFS opposed the choice of CAD disposal, rather than making mere substantive comments, the body of the EIS was the “appropriate point” to disclose and address the views of Dr. Malins and the resource agencies. ONRC v. Marsh, 832 F.2d at 1499, n. 11. Thus, the Corps violated NEPA by failing to expose the uncertainty created by this lack of information concerning polar compounds, the actual toxicity of the “clean” sediments, and the potential chronic effects. See Id. at 1497.

Third, exposing uncertainty and opposing scientific views is not enough, where essential information is lacking. ONRC v. Marsh, 832 F.2d at 1496-97. NEPA requires that the decisionmaker have the information necessary to consider the environmental factors and make a reasoned decision. Half Moon Bay, 847 F.2d at 1394-95; Methow Valley, 833 F.2d at 818. The Memorandum Agreement with NMFS and FWS recognized the need to ensure that the “clean” berm and cap materials are suitable for their intended use, and therefore, required further Microtox toxicity bioassays and chemical analysis for certain pulp mill chemicals. Compl. exh. 5, at 2. However, the Chief Engineer rejected these portions of the Agreement as unnecessary. Compl. exh. 7, at 2, 4-6 ¶ 8. 14 As a result, the Corps lacked accurate, critical information concerning chemical analysis and biological risk.

Chemical analysis did not seek to identify the polar compounds described by Dr. Ma-lins. While the Corps’ scientists claim that polar compounds have not been shown to be toxic, the Tetra Tech/PTI Report and Dr. Malins own studies have shown otherwise. At minimum, the Tetra Tech/PTI Report demonstrates that the polar compounds are toxic and subject to testing, reasons offered by the Corps’ experts for why such tests were unnecessary. The court concludes that testing for polar compounds is possible, and therefore must be conducted before the native sediments can be used for berm and capping material. ONRC v. Marsh, 832 F.2d at 1497. With respect to biological tests, the Corps’ own acute toxicity tests demonstrated that the clean sediments posed significant risks to the environment. If the Corps questions the accuracy of the tests, it should have redone each test with native sediments ob *932 tained in a manner that ensured the efficacy of the biological testing.

Thus, the Corps failed to fulfill its responsibility to gather data and do independent research when faced with uncertainty and gaps in knowledge concerning critically important information. Oregon Environmental Council v. Kunzman, 817 F.2d at 495. The absence of such information “renders it impossible for the [agency involved] to make a reasoned decision.” Me-thow Valley, 833 F.2d at 818. When information essential to a reasoned choice is lacking and obtainable, the agency must obtain it and include it in the EIS. ONRC v. Marsh, 832 F.2d at 1496-97.

In addition, the Corps failed to conduct tests for chronic biological effects due to long-term exposure. Even if the Corps’ experts are correct as to the impossibility of testing for potential chronic effects of contaminants, the Corps may not “simply negate the existence of these impacts.” Methow Valley, 833 F.2d at 817 (citation omitted). If a governmental agency cannot obtain adequate information upon which to make a reasoned assessment of the environmental impacts, it must perform a “worst case” analysis. Id. 15 Consequently, the court concludes that, if tests for chronic biological effects cannot be performed, then the Corps also violated NEPA by failing to perform a worst-case analysis. 16

Finally, the court concludes that separate from the NEPA violations described above, the Tetra Tech/PTI Report constitutes significant new information that requires additional studies and preparation of a supplemental EIS. An agency must supplement its EIS if “[t]here are significant new circumstances or information relevant to environmental concerns and bearing on the proposed action or its impacts.” 40 C.F.R. § 1502.9(c)(l)(ii). “When new information comes to light, the agency must consider it, evaluate it, and make a reasoned determination whether it is of such significance as to require implementation of formal NEPA filing requirements.” ONRC v. Marsh, 832 F.2d at 1494.

An agency determination not to formally supplement an EIS in light of new information withstands judicial scrutiny only if the decision is reasonable. Id. at 1494. Reasonableness depends on such factors as (1) the significance of the new information; (2) the probable accuracy; (3) the degree of care the agency used in considering the information and evaluating its impact; and (4) the degree with which the agency supported its decision not to supplement with a statement of explanation or additional data. Id. In ONRC v. Marsh, the Ninth Circuit found that the Corps violated NEPA in failing to prepare a supplemental EIS when presented with two new studies that, although “not conclusive”, presented a “legitimate concern” about the environmental effects of the Elk Creek Dam. 832 F.2d at 1495.

Here, the court concludes that the Tetra Tech/PTI Report’s identification of polar compounds in East Waterway sediments, and its assertion that such compounds have proved toxic to the biota, is significant new information with a high probability of accuracy. As explained above, the court finds the government’s proffered reasons for not *933 supplementing the record to be nonre-sponsive to the problems raised by Dr. Malins and others. Consequently, the court concludes that the Corps acted unreasonably in failing to conduct supplemental chemical and biological studies given that the Tetra Tech/PTI Report raised “legitimate concerns” that must be investigated, especially in view of the central role the capping material plays in RADCAD’s design.

3. Unconfined Contaminated Sediments

The Corps acknowledges that a certain amount of the contaminated sediments will be released during dredging and disposal (“mass release”). Nevertheless, the FSEIS found it unnecessary to estimate “the magnitude or fate of these minor losses in terms of areal extent or thickness of eventual deposition or potential biologic impacts.” FSEIS, III — 43 — 44; see also ROD, App. A., at 9 (“impacts probably would not be significant”). Subsequent to the publication of the FSEIS, the Corps adopted WQG capping criteria allowing a potentially significant amount of contaminated material to remain uncapped: sediment less than 3 cm thick and 5 per cent of contaminated sediments greater than 3 cm thick. See SHB, Bendor et. al, at 13. Despite the potential for significant ecosystem exposure in addition to the original mass release figures, the Corps adhered to its position that an assessment of the biological impacts was unnecessary. See ROD, App. A, at 9. The court finds that the Corps' handling of the risk posed by uncapped sediments violated its NEPA obligations. NEPA required that the Corps disclose in the FSEIS text that the NMFS had challenged the Corps’ mass release conclusions. NEPA also required the Corps to assess and to disclose the potential biological impacts.

The Corps based its decision not to evaluate the environmental impact of mass released contaminants upon two considerations: (1) the total mass release would be no more than 4.5 per cent (2 per cent of the contaminated sediments would be resus-pended by the dredging action and 2.5 per cent due to release during disposal), Opp. Mem. 30; and (2) any impact from such releases would be “temporary”, FSEIS, I-12, because the existence of a downslope containment berm and the weakness of the bottom currents would assure that the majority of mass released contaminants would settle on the site and be capped. FSEIS, III-43-44.

The court is troubled by the Corps’ heavy reliance on the weakness of bottom currents to support its conclusion that any biological impacts would be minor and temporary. Likewise, the Corps arrived at its 4.5 per cent figure by relying on the weak currents to support its decision not to include an additional “1 per cent to be sheared off the disposal mass during descent, and another 1 per cent to be resus-pended during impact with the sea bottom.” ROD, at 48; see also WES Evaluation, at 132. Yet, the Corps bases its conclusions on a government-commissioned study that measured current conditions at the previously proposed Deep Delta site, and the government has failed to demonstrate that results from this study are applicable to RADCAD. See Northern Technical Services Inc., US. Navy Deep-Delta Confined Aquatic Disposal Site Current Monitoring Study 24 (March 1986) (“Nor-tec Current Study”) (cited at FSEIS at II — 8). Moreover, the results may in fact contradict the Corps’ conclusions because the observed high variability of current direction casts doubt upon the government’s assumption that capping sediments will follow the contaminated sediments’ dispersion pattern. See Nortec Current Study, at 24. Three members of the Shoreline’s Board concluded that while the average bottom currents are unlikely to cause resuspension once the sediments have settled, the surface currents are stronger and “the observed currents, winds, and other physical forces are sufficient to move sediments off-boundary areas.” SHB, Bendor et. al, at 16.

Of even greater concern is the failure of the EIS text to acknowledge the opposing views of the NMFS, FWS, and Dr. Malins concerning the extent and effect of the *934 mass released contaminants. The NMFS opposed the CAD project, in part, because its review of Corps’ documents demonstrated that “mass release of 7 percent of the contaminated material could occur during placement.” Compl. exh. 4, at 7. Government experts contend that the contaminated sediments have been shown to be harmful only under long term, high concentration exposure. See FSEIS, App. S, at 22-23; Anderson Decl., at 17. Nevertheless, after three weeks of scientific testimony, three members of the Shorelines Board stated that “[t]here is no dispute that the surface ‘black mayonnaise’ sediment ... is contaminated” and toxic to harbor marine life. SHB, Bendor et al, at 19. But see PCHB, at 16; SHB, Dufford, et. al (record did not demonstrate that contaminated sediment “would in fact cause acute or chronic toxic conditions to the aquatic biota if disposed without a cap”). Likewise, the NMFS warned that contaminated sediments allowed to remain uncapped “could kill or have sublethal effects on aquatic and fishery resources.” Id. at 7-9. Dr. Malins also warned that sediments like the contaminated layer to be disposed of “are currently associated with chronic disease and even acute lethality in East Waterway biota.” Malins Decl. 1t 22. In his view,

[b]ased on what we do know about the sensitivity of aquatic organisms to these contaminants, the unrestricted release of even 4 % of the presently identified contaminated dredge material (let alone a larger amount) ... will cause biological damage in and around the RADCAD site —and beyond.

Id.

Dr. Anderson further contends that contaminant toxicity will be diluted and degraded by natural processes such as oxygenation and the action of bacteria. Anderson Decl., at 6. While Dr. Malins acknowledges a gap in scientific knowledge as to the intensity of dose and length of exposure required to cause biological harm, he contends that even minute doses of the chemicals present in the sediment can cause biological damages. Malins Decl. 1122. Moreover, as the NMFS points out, the extent of exposure is totally unknown because the capping criteria sets no limit on "the portion of the disposal area where contaminated material is less than 3 cm deep, an area that could encompass several acres.” Compl. exh. 4, at 7; see SHB, Bendor et. al, at 13.

“Where scientists disagree about possible adverse environmental effect, the EIS must inform decision-makers of ‘the full range of responsible opinion on the environmental effects (citation omitted)’ ”. CATS, 428 F.Supp. at 922. Where the agency fails to acknowledge the opinions held by well respected scientists concerning the hazards of the proposed action, the EIS is fatally deficient. Id. The FEIS and FSEIS text failed to disclose the opposition of what must be acknowledged as credible, reliable scientific sources. Here again, the court concludes that based on the circumstances of this case the “appropriate point” to disclose and address these “opposing views” was in the body of the EIS, rather than the comments and response section. See ONRC v. Marsh, 832 F.2d at 1499 n. 11.

The Corps compounded its NEPA violation by failing to perform a worst-case analysis after the WQC capping criteria allowed a potentially significant additional amount of contaminated sediments to remain uncapped. With respect to the WQC criterion allowing for uncapped layers of 3 cm or less, government experts contend that natural processes will cause this layer to be mixed into the existing sediment and thoroughly diluted. Germano Deck, at 21-22. Nevertheless, a significant portion of the contaminated sediments may well remain unconfined because some mass release will travel off-site, remain suspended in the water column, and a significant percentage may be less than the 3 cm-thick capping requirement. Given the preexisting views of credible scientists that even a 4 per cent mass release would cause serious biological impacts, the failure to perform a worst-case analysis cannot be justified. Regardless of the Corps’ view of the likelihood of biological consequences from uncapped contaminated sediments, with the *935 addition of the WQC capping criteria, a worst-case analysis had to be performed. Even where an agency believes that environmental impact is unlikely or improbable, a worst-case analysis must be prepared when there is scientific uncertainty as to the safety of the proposed action. SO-CATE, 720 F.2d at 1476.

4. Potential for Cap Failure

FOE contends that, despite evidence that the cap may be unable to isolate the contaminants, the FSEIS only provides assurances of cap effectiveness. According to FOE, the EISs neither admitted nor assessed the risk. Based on studies conducted by its Waterways Experiment Station (“WES”), the Corps concluded that at minimum an 80 cm cap was required, 30 cm to contain any chemical migration and an additional 50 cm to protect against biotur-bation. FSEIS, III-45. As designed, the cap’s thickness is predicted to average 1.4 m (4.5 feet) over the contaminated materials area. FSEIS, III-45. The 1.4 m figure was arrived at by estimating initial cap thickness at approximately 9 feet and with an assumed settling rate of 50 per cent. Fuglevand Decl. II12. 17 In any event, both the Corps and the WQC set 1 m as the operational requirement. FSEIS, III-45; WQC, at 2 112 (Compl. exh. 1). According to FOE, evidence in the administrative record demonstrates that two natural processes, chemical migration and bioturbation, will likely cause a breach in cap integrity.

a. Chemical Migration

Chemical migration is defined as the process by which chemicals would reach the ecosystem because of water circulation in and around the capped contaminated sediments. WES conducted chemical migration studies for 40 days using two “tracer" chemical compounds, rather than actual contaminants found in the East Waterway sediments. 18 Although the FSEIS cites the WES study, the text of the EIS does not acknowledge that the Corps’ scientists concluded that while capping can isolate the contamination over a period of 40 to 360 days, the capping only slows, but does not prevent chemical migration. FSEIS, App. I to App. A. to App. B., at 1-22. Nor does the EIS text acknowledge the NMFS’ express concern that, in the long-term, cap material would become saturated with chemicals, which would leach into the ecosystem causing adverse effects on aquatic and fishery resources. Compl. exh. 4, at 9.

However, the Corps points out that it addressed these concerns in issuing its Record of Decision. 19 In its summary of comments and responses, the Corps acknowledged that no confinement mechanism will contain the chemicals in the long term, but that thorough on-site monitoring would allow the Corps to take remedial action by providing additional capping material. Compl. exh. 1, App. B (summary of comments received and responses given), at 39-31.

b. Bioturbation

Bioturbation is the redistribution of contaminated sediments into the ecosystem as a result of the active reworking of sediment by burrowing organisms. Germano Decl., at 23; see Suchanek Decl. 11 5. The WES tested actual contaminated sediments covered by a 50 cm cap using worms and clams as test bioturbators. Although the worms burrowed through the cap, there was no significant release of contamination, and the WES recommended that a total cap thickness of 1 m be required: 30 *936 cm for chemical migration, 50 cm for bio-turbation, and a 20 cm margin to allow for irregularities in the cap thickness. FSEIS, III — 45.

Characteristically, the bioturbators used by the Corps do not inhabit the RADCAD cite. Moreover, prior to publication of the FSEIS, the Corps made no attempt to ascertain the existence of native bioturba-tors and conducted no field study of biotur-bation. In addition, the FSEIS text made no mention of the intense criticism its bio-turbation studies engendered in the scientific community. Finally, in March of 1987, in response to concerns expressed by the resource agencies that native bioturbators existed, the Navy commissioned a field study by the University of Washington Fisheries Research Institute. ROD, at 29. The study determined that two bioturba-tors inhabited the RADCAD site: a burrowing shrimp, Axiopsis spinulicauda (“Axiopsis); and a burrowing sea cucumber, Molpadia sp. (“Molpadia”). Dinnel, et. al, Preliminary Survey of Burrowing Infauna in Port Gardner (Final Rep. June 30, 1987) (“Dinnel Survey”), at AR: 38-14.

Very little is known about Axiopsis and Molpadia bioturbation, Suchanek Decl. If 23, but a relative of Axiopsis, Axius ser-ratas, can burrow as deeply as 3 m, “especially in areas of organically rich sediments.” Dinnel Study, AR: 38-8. The Dinnel Survey samplings penetrated to a maximum depth of 2 meters and made only a very limited attempt at gathering samples. 20 Burrows of all sizes were found at the 2 m maximum. AR: 38-14. Other burrows suggestive of Axiopsis activity were found at 60 cm, an actual Axiopsis was found at 43 cm, and the deepest Mol-padia burrow was at the 72 cm. level. The Dinnel Study concluded that (1) “a strong statement cannot be made concerning the burrowing depth of [.Axiopsis and Molpa-dia ]”; (2) the burrows found at the 2 m level “could have been relics of past burrows”; and (3) the potential for burrowing deeper than the 1 m cap requirement was “unknown.” AR:38-20.

The NMFS, the FWS, and Dr. Suchanek conclude that the Dinnel Study did not assess the risk posed to the cap’s integrity by the newly discovered organisms. Compl. ex. 5 112; Suchanek Decl. 111123-32. Rather than address this uncertainty by further studies, as called for in the Memorandum Agreement, the Corps concluded that based on the available data, the predicted 1.4 m average cap thickness would exceed any expected burrowing at RADCAD. ROD, at 30. Further tests were not run because, in the Corps view, “neither of these two creatures may repopulate the site [after disposal and capping] in the same densities.” Opp. Mem. 37. Despite the Corps assurances of no risk, the Corps took account of the potential risk to the cap’s integrity by requiring monitoring of bioturbation at the site for 10 years. ROD at 30. Id.

c. Conclusion: The Threat Needs to be Studied and Exposed

The court concludes that the Corps’ handling of the various threats to cap integrity violated NEPA’s requirement of informed decisionmaking and informed public participation. Half Moon Bay, 847 F.2d at 1394-95; Methow Valley, 833 F.2d at 818. As explained in the context of other technical issues, the body of the Corps’ EIS cannot simply ignore the opposing views of the Federal resource agencies and other credible scientists. ONRC v. Marsh, 832 F.2d at 1499, n. 11. By failing to acknowledge opposing views and the conclusions of its own WES study in the text, and by failing to even test for, let alone use, native bio-turbators, the Corps only provided assurances of cap effectiveness; consequently, the FSEIS reader is apt to conclude that chemical migration and bioturbation posed no threat to cap integrity. Yet, as explained above, the risk is quite real. Thus, the court concludes that the Corps’ FSEIS violated NEPA by neither admitting or assessing the risk despite evidence that the cap may be unable to isolate the contaminants.

*937 The court also concludes that the Corps’ reliance on monitoring and additional capping to address the threat posed by chemical migration and bioturbation does not eliminate the areas of concern arising out of the FSEIS’ NEPA failures. An agency must candidly disclose in its EIS the risks posed by its proposed action. Otherwise the EIS cannot serve its purpose of informing the decisionmaker and the public before the decision to proceed is made. CEQ, Forty Most Asked Questions Concerning NEPA Regulations, Question 34b, Fed. Reg. 18036 (March 23, 1981); see also 40 C.F.R. §§ 1502.1, 1505.2. Even now, the Corps continues to dismiss the possibility that the cap would become saturated by chemicals as being of little concern due to the design of the project and because the diffusion of chemicals through the cap is a lengthy process encompassing centuries. Yet, the Corps' own studies concluded that capping would prevent slow chemical migration over a period of only 40 to 360 days. Given that the court finds the CAD technology experimental at RADCAD depths and the hydraulic capping experimental at any depth, the Corps cannot simply rely on additional capping to avoid assessing the potential environmental consequences and disclosing them to the public and decisionmakers. Given the existing concern with the monitoring plan’s ability to detect problems, 21 it seems inappropriate for the Corps to avoid assessing potential environmental consequences by placing total reliance on the monitoring plan’s ability to detect chemical migration and bioturbation. Clearly the FSEIS violated NEPA requirements by failing to disclose the long-term risk posed by chemical migration and failing to address the potential environmental harm.

Moreover, as the NMFS observed, additional cap material would only postpone the threat from bioturbation and chemical migration, thereby creating a long-term need to add cap material, and making the Corps’ only feasible remedial action no solution at all. Compl. exh. 4, attachment at 4, 7. The continuing need to apply new capping material was never addressed in the Corps’ EIS and, at a minimum, would change the cost of the RADCAD option. Adjustments in RADCAD’s cost would necessarily affect the Corps’ analysis of alternatives. 22

Because the ROD merely relied on monitoring and additional capping to avoid addressing potential environmental consequences, the Corps did not cure its original failure to obtain information necessary to make a reasoned decision. See Half Moon Bay, 847 F.2d at 1395-97.

Finally, the court concludes that, just as in the case of the Tetra Tech/PTI Report, the Dinnel Study constitutes significant new information that required additional studies and preparation of a supplemental EIS. 40 C.F.R. § 1502.9(c)(l)(ii); ONRC v. Marsh, 832 F.2d at 1494-95. Undoubtedly, the Dinnel Study’s identification of native burrowing organisms at the RAD-CAD site is significant new information with a high probability of accuracy. See Id. In fact, the Corps itself recognized the significance of the Dinnel Study, observing that the study produced insufficient data to rule out a danger to the cap’s integrity. AR: 36-2.

The court finds the Corps’ decision not to conduct further studies and not to prepare a supplemental EIS concerning the potential environmental impact from bioturbation to be unreasonable in light of the Dinnel Study’s findings. The Corps decided to address the uncertainty by requiring monitoring for deep burrowing organisms, requiring evaluation of the bioturbation data after each sampling period, and by requiring that such monitoring continue for up to 10 years after the completion of disposal operations. AR:36-2, 4. Yet, the Memorandum Agreement with the NMFS and FWS required the Navy to obtain estimates of burrowing of Molpadia and Ax-iopsis and predict the impact on cap integrity over a ten year period before starting the dredging project. As with other key provisions of the Agreement, the Corps’ Chief of Engineers rejected these require- *938 mente. Compl. exh. 6. Although baseline bioturbation data will be collected, it will not provide answers to the key questions of concern: the depth of the burrows and the rate of redistribution of contaminated material. Compl. exh. 7, at 3-4. As noted above, the Corps now contends that further teste were not run because “neither of these two creatures [Axiopsis and Molpa-dia ] may repopulate the site [after disposal and capping] in the same densities.” Opp. Mem. 37. The record, however, provides no scientific support for this statement. 23 In addition, it is patently unreasonable for the Corps to rely on an experimental technology (CAD) to provide a remedy for a risk of “unknown” dimensions. Moreover, as explained above, little is known about Axiopsis and Molpadia, and if Axiopsis is capable of burrowing to 3 m depths, like its cousin Axius, recapping would not provide a remedy in any event.

In conclusion, the court finds the Dinnel Study to be significant new information within the scope of 40 C.F.R. § 1502.9(c)(l)(ii). The Corps’ determination not to formally supplement its EIS in light of the Dinnel Study cannot withstand scrutiny. Given the accuracy and importance of the Dinnel Study’s discoveries, the court finds that the Corps did not consider and evaluate the information and its impact with the necessary degree of care, nor did the Corps provide sufficient support for its decision not to supplement the record. See ONRC v. Marsh, 832 F.2d at 1494-95. Consequently, the court concludes that the Corps acted unreasonably in failing to conduct supplemental teste to identify the nature and scope of the bioturbation threat at the RADCAD site, as it had agreed to do in its Memorandum Agreement with FWS and NMFS.

C. The Mitigation Plan

Regulations promulgated by the Council on Environmental Quality (“CEQ”) require every EIS to include a discussion of means to mitigate adverse environmental impacts. 40 C.F.R. § 1502.16(h). Adequacy of an EIS hinges, inter alia, on the completeness of the mitigation plan. ONRC v. Marsh, 832 F.2d at 1493; see also, Methow Valley, 833 F.2d at 819. 24 Here, the Corps EIS discusses various mitigation measures. FSEIS, XII-05-14. With reference to the CAD proposals, it is clear that the monitoring plan is the centerpiece of the Corps’ mitigation plan. The government repeatedly relies on the monitoring plan to dismiss the views of the FWS, the NMFS, Dr. Malins, and others concerning the myriad of uncertainties surrounding the CAD technology and the threat to the ecosystem posed by the clean sediments, bioturbation, chemical migration, and unconfined contaminated sediments. See, e.g., ROD at 30 (unknown risk to the cap’s integrity due to native bioturbators); FSEIS, App. B, at 15 (potential environmental risk from failure requires effective monitoring plan). 25 Because the Corps relied on the existence of a monitoring plan to avoid resolving uncertainties or evaluating risks, the need for an adequate *939 monitoring plan cannot be overemphasized. The court has great concern with the process used to develop the monitoring plan and with the plan’s ability to actually detect and prevent against environmental harm.

1. Development of the Monitoring Plan

The Corps’ process for developing the monitoring plan clearly violated NEPA. An EIS “must analyze the mitigation measures in detail and explain the effectiveness of the measures.” Methow Valley, 833 F.2d at 819; ONRC v. Marsh, 832 F.2d at 1493; Northwest Indian Cemetery Protective Ass’n v. Peterson, 795 F.2d 688, 697 (9th Cir.1986), reversed on other grounds, — U.S.-, 108 S.Ct. 1319, 99 L.Ed.2d 534 (1988). Here, the FSEIS text devotes a single paragraph to dredging monitoring, explaining that “no attempt has been made to identify [monitoring plan] ‘trigger’ levels ... [because]_ [t]hese levels will be determined in part by the regulatory agencies responsible for water quality.” FSEIS, XII-8. The Corps did not attempt to craft a detailed monitoring plan during the EIS stage, but rather set out some draft requirements and allowed the Navy to develop a comprehensive plan in conjunction with the appropriate regulatory and resource agencies. FSEIS, XII-9. However, reliance on a detailed plan to be developed in the future with other agencies, state or federal, does not meet an agency’s NEPA obligations. ONRC v. Marsh, 832 F.2d at 1493.

Although the text does not reference it, there is a technical appendix to the FSEIS containing “draft monitoring plans.” FSEIS, App. I to App. A of App. B, at 1. The Corps argues that the draft monitoring plans satisfied its NEPA obligations. The court disagrees. The FSEIS recognized the plans’ limitations, stating that they “cannot be considered final and must be refined once final scheduling and design for the project has been completed.” FSEIS, App. I to App. A of App. B, at 1. Moreover, even though the draft monitoring plans are more detailed for CAD, the court finds that they constitute no more than generic outlines identifying the type of data to be monitored, and make no attempt to set performance or other operational standards. The FSEIS discussion is conclusory and general, and lacks both a detailed description of mitigation measures and analysis of their effectiveness as required by NEPA. Methow Valley, 833 F.2d at 819.

The Corps argues further that, even if a full monitoring plan should have been included in the FSEIS, the NEPA violation has been cured because a detailed monitoring plan now exists. However, timeliness of the mitigation plans’ preparation is essential: “subsequent preparation of an elaborate mitigation plan ‘would not absolve a failure to prepare an adequate EIS.’ ” Methow Valley, 833 F.2d at 819. Where an EIS fails to contain a detailed mitigation plan, the agency fails to meet its touchstone obligation of fostering informed decisionmaking and informed public participation. Here, the Corps published the FSEIS in November of 1986, but no detailed monitoring requirements existed until the WQC issued in March of 1987. The Navy issued its draft monitoring plan in July of 1987, but did not finalize the plan until November, two months after the Corps issued the 404 permit. At no time between November of 1986 and November of 1987 did the Corps take public comment on the adequacy of the WQC requirements or the Navy’s monitoring plan.

Although the public did not have an opportunity to comment on monitoring plan specifics during the Corps’ post-FSEIS process, the government contends that nothing further can be accomplished by sending the Homeport EIS back for further review on the basis of an incomplete mitigation plan. According to the government, the plan has been thoroughly aired through the state’s extensive public process. See Thornton Decl. In addition, the government contends that various agencies, such as the NMFS and the FWS, had ample opportunity to analyze and scrutinize the monitoring plan and share their criticisms with the *940 Corps and Ecology. See Compl. exhs. 7, 8, 10, 11, and 13.

The court doubts that the Corps can fulfill its own obligations to foster informed public participation by relying on the state’s WQC public comment periods. 26 By relying on the State permit process, the Corps attempts to unlawfully escape its responsibility to fully address the key mitigation measure. See, e.g., Monongahela Power Co., 809 F.2d at 53 n. 114 (requirement of a section 401 permit not intended in any way to supplant requirements for obtaining a 404 permit); Save Our Wetlands, Inc. v. Sands, 711 F.2d 634, 641-43 (5th Cir.1983) (NEPA requires the Corps to independently evaluate a state or other third party’s findings concerning environmental consequences); Sierra Club v. Alexander, 484 F.Supp. at 464 (state administrative body action cannot not relieve the Corps of its NEPA obligations).

In any event, the State’s process did not cure the problem. Although the State conducted a public hearing in January of 1987, the all important Appendix B of the WQC did not exist until its publication in March. The State limited further public comment to whether the Navy’s draft plan met the State-imposed requirements. Because of the limited nature of the State public comment period, and the lack of any Corps public process, the public was not aware of the NMFS’ and the FWS’ criticism or the Corps’ own criticisms of the monitoring requirements. The failure to adequately describe and address the monitoring plan in the EIS resulted in the public never having the opportunity to review the plan, comment on it, or receive the benefit of the Navy’s response to criticisms of plan lodged by NMFS, FWS, and the Corps itself. Thus, with respect to the monitoring plan's process, the Corps violated NEPA by failing to foster informed decisionmaking, including informed public participation.

2. Adequacy of the Monitoring Plan

FOE also contends that the monitoring plan is incapable of detecting, and preventing against, environmental harm. Without doubt, the final monitoring plan is extremely detailed. See AR:44-1 et. seq; SHB, Dufford et. al, at 30 (finding that the monitoring plan exceeds that applied to any other disposal project anywhere). In issuing the 404 permit, the Corps concluded that the State monitoring requirements “are adequate to properly evaluate the effectiveness of the dredging/disposal operations and to allow for any subsequent determinations of what remedial action will be required if capping is not as effective as proposed.” ROD, at 62. Nevertheless, the court is persuaded that the plan may not effectively detect or prevent environmental harm during RADCAD’s various phases or beyond.

The FWS, NMFS, and members of the Corps itself, have severely criticized several areas of the plan, and the government’s responses do not engender great confidence. 27 In fact, the FWS, NMFS, and three members of the Shoreline Board view the criteria as incapable of protecting against the dangers recognized in the Corps’ EIS and ROD. See compl. exh. 4; SHB, Bendor et. al, at 32-33. For example, the monitoring plan sets out physical performance criteria to determine whether the Navy has successfully disposed of and capped the sediment. As explained earlier, although the Memorandum Agreement established a performance criteria requiring that a minimum of 90 per cent of Phase I contaminated sediments be placed in the Navy’s identified boundaries, the Chief Engineer removed this requirement, and the WQC criteria declares Phase I a success even if contaminated sediments spread to 250 per cent of the boundary. See Compl. exhs. 5, 6, and 7. Other examples of al *941 leged deficiencies in physical criteria include: (1) testing for the 1 m cap thickness-requirement within 30 days of Phase I completion, even though the Corps predicted a 50 per cent consolidation rate over the first year, ROD, ex.B, at 56; 28 (2) no limit exists on the amount of area that can be covered by uncapped sediments less than 3 cm thick; (3) Phase I, which involves only Vs of the contaminated sediments, will have succeeded even if contaminated sediments 3 cm thick reach the Phase II overall boundary; and (4) no pass/fail limits exist as to the total amount of contaminated sediments less than 3 cm thick that can fall outside any boundary. See SHB, Bendor et. al, at 32-33.

The second area of controversy concerns the lack of biological performance criteria in the monitoring plan. While biological data will be gathered, no performance standards exists to measure the environmental effects of the dredging and disposal. At least one member of Corps commented that

[t]he biological portions of the monitoring plans violate the most basic precepts of any monitoring activity; in essence, they are merely a data gathering activity. Above all, a monitoring program must have management objectives, criteria which determine whether or not the objectives are being met, and a means of remedial action to achieve the objectives if they are not met.

Letter of Warren E. Baxter, Chief, Regulatory Branch, Dep’t of Army, to Washington Department of Ecology, Aug. 27, 1987, Ex. E-4, at 4 (compl. ex. 11); Letter of NMFS, 10/5/87 ¶ 15 (compl. ex. 7). Despite this internal criticism of the lack of biological performance standards, the Corps issued the Navy a 404 permit based on the draft monitoring plan. In response, the government cites Dr. Anderson, who contends that there is no standard biological test for chronic conditions and any such test would take years to run. Anderson Deck, at 17-18. The Corps further argues that few biological tests have been accepted as reliable and few tests show causal connections between contaminants and mortality or disease. Opp.Mem. 55. As explained earlier, the absence of available tests does not relieve the Corps of its responsibility to perform a worst case analysis.

Finally, the NMFS, Dr. Malins, and Dr. Suchanek criticize the biological data that will be collected, characterizing it as incapable of revealing changes in the Port Gardner ecosystem as a result of the Navy dredging. For example, the NMFS contends that because of the time constraints imposed by the homeport construction schedule, any baseline data developed will be inadequate, making it impossible to distinguish significant changes from natural variation. Letter of Dec. 17, 1987, compl. exh. 13, at 2. Dr. Suchanek criticizes the baseline monitoring for bioturbation as inadequate, contending that no statistical tests can be run on the limited sampling that will occur and that sampling will only penetrate to the 50 cm level, rather than the necessary 2-3 m level. Suchanek Decl. ¶ 34-36. Dr. Malins contends that the bio-accumulation studies are so limited that the entire spectrum of the disposal site aquatic community will not be adequately represented. Malins Deck ¶ 49. In response, the Corps cites the fact that data collection will continue to monitor the benthic community for 10 years. See Germano Deck, at 27. Nevertheless, the WQC contains no biological pass/failure criteria and the Memorandum Agreement with the NMFS and the FWS recognized that sampling of a single benthic species was insufficient; at least three species had to be sampled to assess hioaccumulation in a “statistically significant manner.” Compl. exh. 7, at 6, 1115.

3. Conclusion

Because the court has found various NEPA violations with respect to the technical issues, violations that can be cured only by additional testing and/or a supplemental EIS, the court need not reach the question whether the Corps’ treatment of the moni *942 toring plan would have led independently to an injunction. However, the Corps must cure its flawed public process. It can do so as part of its supplemental process, by completely disclosing agency criticism of the monitoring program and by taking public comment on the sufficiency of the plan. Likewise, when the Corps acts to supplement the record, it must specifically analyze and respond to the views of the resource agencies and other scientific opinions as to the sufficiency of the various monitoring plan requirements. Because of the crucial role assigned to the monitoring plan, the Corps must seek to ensure that the plan has the ability to detect and prevent against environmental harm. Consequently, because the court has found the CAD technology to be experimental at best, the court strongly suggests that the Corps adhere to the Memorandum Agreement requirements deemed necessary by the NMFS and FWS “to determine the effectiveness of untried deep water disposal process.” Letter of Sept. 18, 1987, AR 12:2; see Compl. exhs. 5 and 7.

D. Treatment of Reasonable Alternatives

Under NEPA, an EIS must include alternatives to the proposed action. 42 U.S.C. §§ 4332(C), 4332(2)(E). The CEQ characterizes the discussion of alternatives as “the heart of the environmental impact statement.” 40 C.F.R. §§ 1502.14,1502.16. To withstand review, an EÍS must consider every reasonable alternative. Methow Valley, 833 F.2d at 815. Here, the issue is not whether the Corps and Navy considered all reasonable alternatives; rather, the question is whether the Corps obtained sufficient data to make a reasoned choice between CAD and various upland alternatives. A decisionmaker must explore alternatives in sufficient enough detail to “sharply defin[e] the issues and provid[e] a clear basis for choice among options by the decisionmaker and the public.” 40 C.F.R. § 1502.14. Here again “the touchstone” of the court’s inquiry “is whether an EIS’s selection and discussion of alternatives fosters informed decision-making and informed public participation.” State of Cal. v. Block, 690 F.2d at 767.

Among other alternatives, the Corps and the Navy explored two types of upland disposal plans at Smith Island, located four miles upstream from the dredging area. One of the alternatives, “excavated” disposal, requires construction of a dike and excavation of a “cell” below the ground water level. An ‘elevated’ disposal plan requires construction of a perimeter dike above the existing ground level and water table. FSEIS, App. C., at 4,5. The government acknowledges that both Smith Island alternatives were found to be feasible. See FSEIS, III — 85 — 104; Appendix C.

At the outset, the court notes that the choice of RADCAD as the preferred alternative hinged on the Corps’ conclusion that the Smith Island alternatives cost significantly more and that RADCAD equaled these upland options in terms of technological feasibility, risk, monitoring, and environmental considerations. FSEIS, III — 107— 11; ROD, Appendix A, AR: 13-162, 13-253-56. Putting aside the question of the accuracy of cost estimates, additional cost would be determinative only if the Corps could support its conclusions as to the equivalency of technological feasibility, environmental risk, and ability to monitor. However, given the problems already described with respect to the Corps’ analysis of RADCAD’s potential environmental impacts, the court finds the Corps’ analysis of alternatives and choice of RADCAD as the preferred alternative to be severely deficient.

For example, the choice of RADCAD as the preferred alternative hinges on the Corps’ belief that CAD presents technological risks equal to the upland disposal systems. There is no doubt that the upland disposal options pose some environmental risk. See SHB, Bendor et. al, at 34-35. However, the FWS, the NMFS, and FOE’s expert, Dr. Jerry Ongerth, offered persuasive evidence that the upland disposal sites present fewer negative environmental impacts and carry fewer long term risks than RADCAD. NMFS Letter of Apr. 3, 1987, at 6 ¶ 6 (compl. ex. 4); Ongerth Decl. 1121. As Dr. Ongerth points out, “upland waste disposal technology is standardized, *943 established, and of proven reliability,” On-gerth Deel. II21, as opposed to RADCAD, which the court finds both experimental and unproven.

With respect to environmental considerations, the Corps’ rating of RADCAD as equivalent to the upland options cannot withstand scrutiny because the court has already found the FSEIS’ analysis inadequate as to the environmental risk posed by unconfined contaminated sediments, clean sediments, bioturbators, and chemical migration. The Corps defends its choice of RADCAD by characterizing these potential impacts as not significant, AR:13-256, and relies on the monitoring plan to cure any “potential major environmental impacts.” Id. But as already explained, great uncertainty exists concerning the monitoring plan’s ability to detect and take action against environmental injury. By contrast, unlike the complexities of deep water monitoring, upland monitoring technology is well established and non-experimental, and correction of potential problems is easily manageable due to the accessibility of the upland site. Ongerth Decl. 1118. Dr. On-gerth’s views are fully supported by the NMFS and FWS, which concluded that, with the virtual abrogation of the Memorandum Agreement, upland disposal was far superior, inter alia, because the Navy’s monitoring program cannot determine CAD’s effectiveness. Letter of Sept. 18, 1987, AR 12:2.

Given the above considerations, the Corps’ choice of RADCAD cannot be a reasoned selection among alternatives. Clearly the Corps lacked information that is “ ‘important,’ ‘significant,’ or ‘essential’ to a reasoned choice among alternatives”. See Oregon Environmental Council v. Kunzman, 817 F.2d at 495. In the absence of such information, a choice among alternatives cannot be said to be reasonable.

FOE also challenges the Corps’ estimated costs of the various alternatives and the Corps’ failure to conduct further tests on the Smith Island alternatives. Undoubtedly, the cost differential was the key factor in the rejection of upland alternatives. FOE’s challenge to the accuracy of the Corps’ cost figures will be discussed infra. 29 However, RADCAD’s estimated costs must be revised to reflect, inter alia, the likelihood that the risks posed by bio-turbation and chemical migration may necessitate repeated cap augmentation long into the future. For the Corps to make a reasoned choice among alternatives, costs of additional future capping must be included in RADCAD’s cost estimate.

The court rejects FOE’s contention that the Corps had to conduct additional Smith Island studies to make a reasoned choice. The FSEIS devotes 17 pages to discussion of the Smith Island options and Appendix C contains a detailed feasibility design study, which concludes that the options are feasible but “will require collection of additional site soils and groundwater data, in addition to laboratory testing and in-depth engineering analysis.” FSEIS, App. C, at 15. The Corps established sufficient criteria to make a reasoned choice, criteria that could be evaluated for the upland sites without the need for additional studies. See FSEIS, IH-107-11. However, the court agrees with NMFS that a reasoned decision requires the addition of criteria for risks due to operational mishaps and the capability for accurate monitoring. See FSEIS, XIII-55. Moreover, if the Corps chooses again to compare alternatives using a matrix displaying numerical ratings, see FSEIS, Ill-Table 3-9, 3-10, the Corps must explain how it arrived at the numerical ratings assigned to each alternative for each criterion. In addition, if factors other than those displayed on the matrix play a role in the decision, see FSEIS III — 111, the Corps must present a clear, cogent comparison of the alternatives based on these additional factors. Finally, if the NMFS continues to severely criticize the methodology used to rank the various alternatives and the actual rankings assigned to the alternatives, see FSEIS, XIII-55, the Corps must disclose the NMFS’ views in any subsequent supplemental EIS.

*944 E. Summary: NEPA Violations

The court concludes that with respect to each of the three major areas of concern, the Corps and the Navy failed to satisfy their NEPA obligations because (1) they did not provide “an environmental disclosure sufficiently detailed to aid in the substantive decision whether to proceed with the project in light of the environmental consequences,” and (2) they failed to provide “the public with information and an opportunity to participate in gathering information.” Methow Valley, 833 F.2d at 814. Thus, in reexamining the environmental problems surrounding the Navy’s proposal and in redetermining whether to issue the Navy a 404 permit, the Corps must carefully adhere to public comment requirements and address the following: 30

a. Technological Feasibility:

(1) disclose in the text of any supplemental EIS the experimental nature of and the uncertainty surrounding CAD and RAD CAD, including the basis of the opposing views of the NMFS, FWS, Dr. Malins, and other credible scientists; and

(2) assess and disclose the “major” environmental impact of a complete or partial failure of the CAD technology during both Phase I and Phase II of the project. On this point the supplemental EIS must be particularly thorough because the environmental consequences of the proposed action are great.

b. Sediment Characterization of Berm and Cap Materials:

(1) conduct further chemical analysis to identify chemicals in the “unresolved complex mixture” and to test for polar compounds found in the Tetra Tech/PTI Report;

(2) conduct further biological acute toxicity tests on organisms other than amphipods;

(3) conduct tests for chronic biological tests, but if such tests can be shown to be too costly or impossible to perform, prepare a worst-case analysis; and

(4) reflect in the text of any supplemental EIS the views of the NMFS and FWS on the above test results.

c. Unconfined Contaminated Sediments:

(1) assess and disclose the amount of mass released contaminants that would move off site due to the combined action of surface and bottom currents, winds, and other physical forces; and

(2) perform a worst-case analysis of the potential biological impact result ing from the total amount of unconfined contaminants from all sources 31 in case of a technological success as defined by the WQC criteria or in the case of technological failure.

d. Bioturbation and Chemical Migration:

(1) assess and disclose the long-term risk posed by chemical migration and the potential worst-case environmental harm; and

(2) conduct further statistically relevant studies to obtain estimates of the burrowing capacity of Axiopsis and Molpadia, the rate they redistribute contaminated material, and predict their impact on cap integrity over a ten year period.

e. Monitoring Plan:

(1) in supplementing the record, specifically analyze and respond to the views of the resource agencies and other scientific opinions as to the sufficiency of the various monitoring plan requirements. As the court suggested, adherence to the terms of the Memorandum Agreement would be the best method for ensuring that the plan has the ability to detect and prevent against environmental harm;

*945 (2) as part of its supplemental process, the Corps must completely disclose agency criticism of the monitoring program and take public comment on the sufficiency of the plan,

f. Reasonable Alternatives:

(1) after reexamining the environmental concerns set forth above with respect to technological feasibility, unconfined contaminated sediments, clean sediments, bioturbators and chemical migration, and the monitoring plan, the Corps must present a clear, cogent comparison of the alternatives;

(2) that comparison must

(a) include additional criteria for risks due to operational mishaps and the capability for accurate monitoring;

(b) explain how various numerical ratings were assigned to each alternative for each criterion and explain in detail what factors other than those displayed on the matrix played a role in the decision; and

(3) revise RADCAD’s estimated costs to include additional costs identified in this opinion including, but not limited to, costs for future capping to meet the risk of bioturbation and chemical migration.

Of course, if the Corps can demonstrate that any of the additional tests the court finds to be necessary can not be performed due to feasibility concerns or exorbitant costs, then a supplemental worst-case analysis would be required. See Methow Valley, 833 F.2d at 817 (citation omitted). In addition, to the extent that the NMFS and the FWS continue to oppose CAD and RADCAD, the Corps must reflect the resource agencies’ views on particular issues at each appropriate point in the text of the supplemental EIS. Likewise, other credible opposing scientific views must be acknowledged and disclosed in the text when appropriate.

CLEAN WATER ACT — SECTION 404 PERMIT DECISION

FOE challenges the Corps’ decision to grant the Navy’s 404 permit application on several grounds. To issue a 404 permit, the Corps must apply EPA’s criteria set out in 40 C.F.R. § 230.10 et seq. Here, FOE contends that the Corps violated section 404 of the CWA by failing to demonstrate that its decision satisfied EPA guidelines with concern to significant degradation, 40 C.F.R. § 230.10(c), and practicable alternatives, 40 C.F.R. § 230.10(a). In addition, FOE contends that the Corps violated its own public comment regulations, 33 C.F.R. § 325.3(a).

A. Standard of Review

Review of the Corps decision to grant a 404 permit also comes under section 706(2) of the APA, “which provides that the reviewing court shall set aside any agency action found to be ‘arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.’ ” FOE v. Hintz, 800 F.2d at 831. This is a deferential standard of review, and “[t]he court may not set aside agency action as arbitrary or capricious unless there is no rational basis for the action.” Id. Nevertheless, the court must carefully review the agency’s explanation for its action and must “consider whether the decision was based on a consideration of the relevant factors and whether there has been a clear error of judgment.” Motor Vehicle Mfrs. Ass’n v. State Farm Mut., 463 U.S. 29, 43, 103 S.Ct. 2856, 2867, 77 L.Ed.2d 443 (1982).

B. Significant Degradation

EPA Guidelines prohibit the Corps from issuing a 404 permit that “will cause or contribute to significant degradation of the waters of the United States,” as determined by appropriate factual determinations, evaluations, and tests. 40 C.F.R. §§ 230.10(c), 230.11. Because the Corps’ ROD relies on the EISs and the studies cited therein to conclude that the RADCAD project will not cause significant degradation, FOE contends that the Corps acted contrary to law and arbitrarily and capriciously. The government contends that the Corps conducted a thorough and accurate analysis of RADCAD’s environmental im *946 pacts and that its factual findings show that no significant environmental harm will occur.

Given the court’s findings and conclusions with respect to FOE’s NEPA challenge, the court must set aside as arbitrary and otherwise not in accordance with law the Corps’ finding that RADCAD would not result in any significant degradation. Contrary to the government's contention, the data gaps and scientific uncertainty surrounding CAD and RADCAD cause the court to conclude that the Corps has not demonstrated the following: (1) the capping process itself poses no environmental risk because it will work effectively; (2) the clean sediments will cause no harm to the biota; (3) the amount of contaminated sediment released will be minor and the effect of that release will be minor; (4) neither chemical migration nor bioturbation will seriously compromise the cap’s integrity; and (5) monitoring of the disposal will insure against any significant cap failure. Because the Corps based its determination of no significant degradation on its assertion that each of the above had been demonstrated, the court finds that the Corps did not sufficiently consider the relevant factors and has committed a clear error of judgment. Motor Vehicle Mfrs. Ass’n., 463 U.S. at 43, 103 S.Ct. at 2867. Without its underlying rational basis, the Corps’ finding of no significant degradation must be considered arbitrary, and the 404 permit must therefore be set aside. See FOE v. Hintz, 800 F.2d at 831.

C. Practicability of the Upland Disposal Alternative

The 404 guidelines prohibit the Corps from issuing a dredge and fill permit under section 404 if “there is a practicable alternative to the proposed discharge which would have less adverse impact on the aquatic ecosystem so long as the alternative does not have other significant adverse environmental consequences.” 40 C.F.R. § 230.10(a). In 40 C.F.R. § 230.10(a)(2), the guidelines define a “practicable alternative” as one that

... is available and capable of being done after taking into consideration cost, existing technology, and logistics in light of overall project purpose.

In issuing the 404 permit, the Corps stated “only RADCAD is considered to be a practicable alternative, in view of costs, existing technology, and logistics.” ROD, at 29. As explained in the NEPA discussion, the Corps reached this conclusion primarily by arbitrarily considering the upland alternatives and RADCAD equivalent on all bases except for cost. In fact, the ROD states that if cost differences were ignored, the Corps could demonstrate that “disposal at RADCAD would not have greater adverse impact on [water] resources than would the Smith Island alternatives.” AR: 13-256 1110.1. Whether RADCAD remains a practicable alternative depends upon the results of the further investigations and worst-cast analyses the Corps must perform to cure the NEPA violations. Thus, because the Corps’ decision no longer has a rational underpinning, the court must set aside the Corps’ choice of RADCAD as the only practicable alternative. See FOE v. Hintz, 800 F.2d at 831.

FOE also challenges the specifics of the Corps’ finding of the Smith Island alternatives as impracticable. Although the ROD did not expressly state why the Corps found the Smith Island alternatives impracticable, it seems clear from the 404 Evaluation attached to the ROD that cost was the primary basis for this finding. See ROD, App. A, at 29, 30, 32. FOE contends that the Corps acted arbitrarily and capriciously in relying on cost to characterize the upland alternatives as impracticable. In fact, the Corps failed to explain why a cost differential of whatever size renders the upland alternatives impracticable. The Corps does not explain why additional funds would not be available. Implicit in the ROD is the Corps’ conclusion that the sheer difference in the cost of RADCAD and Smith Island alternatives, taken in the context of the Home-port's overall budget, demonstrates why the upland alternatives would be impracticable. Undoubtedly, additional cost can constitute a bona fide reason for classify *947 ing an alternative as not practicable. See, e.g., FOE v. Hintz, 800 F.2d at 833-34. (finding it rational that, due to additional costs, two alternatives were not practicable for the applicant). However, significant additional cost can prove determinative, in and of itself, only if the competing alternatives can reasonably be viewed as equivalent with respect to other factors. Consequently, whether the alleged cost differential makes the upland options impracticable depends upon a comparison of RADCAD and the Smith Island alternatives with respect to technological feasibility, potential for environmental harm, ability to be monitored, and potential for remedial action. Until the Corps meets its NEPA obligations, no cost-benefit analysis can be undertaken for either alternative.

The court rejects most of FOE’s arguments concerning the accuracy of the Corps’ cost estimates. Thus, the court concludes that the Corps properly viewed the costs in the context of a $272 million overall Homeport budget 32 and acted rationally in excluding speculative possibilities for a resale of Smith Island in the future. 33 Likewise, the court finds that the Corps rationally included potential additional costs in its revised figure for the upland excavated option. 34 However, as the court indicated earlier, RADCAD also will entail potential additional costs in the form of repeated applications of additional capping material necessitated by chemical migration and bioturbation. These additional costs must be included in RADCAD’s cost estimates. Likewise, the costs of monitoring the RADCAD site, at minimum for 10 years, and probably far longer, must be included. The Corps cannot chose to ignore monitoring costs because it “assumes” that monitoring costs for the Smith Island site would be equivalent. See ROD, App. A, at 30. In fact, nothing in the record supports the Corps’ assumption. 35

D. Public Comment on the Monitoring Plan

FOE contends that the process used to develop the monitoring plan violated the Corps own regulations governing public comment. The court agrees. The Corps’ public notice provision recognizes public comment as essential to the permitting process, and requires that the notice “include sufficient information to give a clear understanding of the nature and magnitude of the activity to generate meaningful comment.” 33 C.F.R. § 325.3(a). Under this regulation, the Corps must “present for public scrutiny the rationale and pivotal data underlying its proposed action before the close of the comment and hearing.” National Wildlife Federation v. Marsh, 568 F.Supp. 985, 994-96 (D.D.C.1983) (because report discussing new alter natives and a cost benefit analysis was published after the close of the comment period, court set aside 404 permit, finding that Corps violated its public comment requirement because new data “was not simply a ‘bit of background information’ ”). 36

*948 As the court has already explained in the context of NEPA’s requirements, a simple review of the time line in this case demonstrates that the Corps did in fact violate its own regulations, in that the Navy published its final monitoring plan in November of 1987, and the Corps approved it in April of 1988, nearly sixteen months after the close of the public comment period. 37 Nevertheless, the Corps makes several arguments to support the lawfulness of the process.

The Corps maintains that 33 C.F.R. 325.-3(a) lists what the public notice must include and none of the items even remotely involves information contained in the monitoring plan. While section 325.3(a) does contain a list, it is by no means exclusive, and the analysis in National Wildlife is persuasive: without pivotal data and information, public comment cannot be meaningful. Next, the Corps argues that the monitoring plan was not “pivotal data underlying the Corps' action.” As the court observed earlier, this contention contradicts the EISs and the ROD itself. 38 Finally, the Corps argues that the State of Washington’s public process on the WQC provided the public with a full opportunity to comment on the monitoring requirements and the public had a similar opportunity to comment at meetings sponsored by the Corps. However, as the court already explained, the Corps cannot escape its independent obligations by relying on the State’s process and, in any event, the State’s public comment process did not cure the problem. 39

The process used by the Corps, in effect, prevented the public from commenting on the single most important feature of the RADCAD project — the monitoring plan. As the court observed in its discussion of NEPA, the public was not aware of the NMFS and the FWS criticism or the Corps’ own criticisms of the monitoring requirements. The failure to include pivotal data in the public notice and to provide a subsequent public notice resulted in the public never having the opportunity to review the monitoring plan, comment on it, or receive the benefit of the Navy’s response to criticisms of the plan lodged by NMFS, FWS, and the Corps itself. Consequently, the court concludes that, just as in National Wildlife, “inclusion of [pivotal data] in the administrative record after the close of the comment and hearing period had the effect of shielding the essential data and the agency’s rationale from public hearing and comment.” National Wildlife, 568 F.Supp. at 994.

APPROPRIATENESS OF A PERMANENT INJUNCTION

In determining whether to issue an injunction where statutory violations have occurred, the court must engage in a two part analysis. See Amoco Prod. Co. v. Village of Gambell, Alaska, 480 U.S. 531, 107 S.Ct. 1396, 1402-04, 94 L.Ed.2d 542 (1987); Save the Yaak Committee v. Block, 840 F.2d 714, 722 (9th Cir.1988). First, the court must determine whether the statute restricts the court’s equity jurisdiction, that is whether the statute would either require or preclude issuance of an injunction to remedy a violation. Village of Gambell, 107 S.Ct. at 1403; Weinberger v. Romero-Barcelo, 456 U.S. 305, 313, 102 S.Ct. 1798, 1803, 72 L.Ed.2d 91 (1982). If the court finds that no such limitation can be found either in the statutory language itself or as a necessary and inescapable inference of the statutory text, then the court must engage in traditional equity balancing to determine the appropriateness of an injunction. Village of Gambell, 107 S.Ct. at 1402; Save the Yaak, 840 F.2d at 722. In the instant circumstances, the court concludes that both the language of the CWA and the results of traditional equity balancing require a permanent injunction to ensure compliance with statutory mandates.

It seems well established that violations of NEPA, in and of themselves, do not necessitate an injunction to ensure compli- *949 anee; nothing in NEPA indicates that Congress intended to limit a court’s equity jurisdiction. Save the Yaak, 840 F.2d at 722. However, the court finds otherwise in the case of the CWA.

Recognizing that a court must not lightly assume that Congress intended to require an injunction, see Village of Gambell, 107 S.Ct. at 1402-03, the court nevertheless concludes that the CWA contains a necessary and inescapable inference that an injunction must issue to ensure compliance when a failure to enjoin the proposed action would introduce a pollutant into the Nation’s water. In 38 U.S.C. § 1311(a), Congress expressly mandates that “the discharge of any pollutant by any person shall be unlawful” unless there has been compliance with a variety of statutory provisions including issuance of a 404 permit pursuant to 33 U.S.C. § 1344. Here, the court has set aside the Navy’s 404 permit, inter alia because the Corps could not conclude that the Nation’s waters would not be degraded by the dumping of highly contaminated sediments and allegedly clean sediments, which may in fact be contaminated. A failure to enjoin the Navy’s disposal project would subject the Nation’s waters to a real and present danger of pollution. This aspect distinguishes the instant situation from Romero-Barcelo. There, the Supreme Court held that the Federal Water Pollution Control Act did not require an injunction of the Navy’s discharging of artillery shells into the United States waters because the statute’s purpose would not be undermined by allowing a statutory violation to continue where no pollution was occurring. Village of Gambell, 107 S.Ct. at 1403 (explaining Romero-Barcelo, 456 U.S. at 314-15, 102 S.Ct. 1804). Likewise, the real danger of the pollution that would occur absent a court injunction distinguishes this case from Village of Gambell. There, the Supreme Court held that where a statutory violation does not undermine the substantive policy contained in the statute, an injunction does not automatically issue. Village of Gambell, 107 S.Ct. at 1403.

Even had the court concluded that the CWA does not require an injunction to ensure compliance in this situation, the court concludes that the nature of the NEPA and CWA violations in this instance necessitates an injunction under the well established equity principles governing injunc-tive relief. “The basis for injunctive relief is irreparable injury and inadequacy of legal remedies.” Village of Gambell, 107 S.Ct. at 1402. “In each case, a court must balance the competing claims of injury and must consider the effect on each party of granting or withholding of the requested relief.” Id. Where an agency fails to thoroughly evaluate the environmental impact of a proposed action, there is no “presumption” of environmental injury. Village of Gambell, 107 S.Ct. at 1404. However, because environmental injury is by its very nature irreparable, when environmental injury is “sufficiently likely, the balance of harms will usually favor the issuance of the injunction to protect the environment.” Id.; Save the Yaak, 840 F.2d at 722.

Here, the court has concluded that there is a clear risk of environmental injury from a variety of sources including a technological failure, the “clean” sediments, the unconfined contaminated sediments, and threats to the cap’s integrity. In addition, it is uncertain whether monitoring of the disposal will insure against the likelihood of any of these risks from occurring. Because the Corps failed to assess chronic biological effects, failed to properly test the “clean” sediments, and failed to conduct necessary investigations into other issues such as bioturbation, it is impossible to ascertain the exact likelihood of environmental injury. The court can state, however, that a significant risk of major irreparable environmental impact exists if the dredging proceeds as planned. Although the court accepts the Navy’s assertion that a delay in the project will set back its national security plans, the court concludes that a delay in the Everett Homeport will not irreparably injure the nation’s defense capabilities. In fact, Congress expressly provided that no funds be obligated or expended for the Homeport until all environmental considerations were fully addressed. See National Defense Authoriza *950 tion Acts of 1987, Pub. L. No. 99-661, § 2207 (1986); National Defense Authorization Acts of 1988, Pub.L. No. 100-180, § 2322 (1987); FOE v. Hall, NO. C88-380R, Order Granting Plaintiffs’ Motion to Clarify (August 8, 1988) (“Clarification Order”).

After balancing the competing claims of injury and considering the effect on each party of granting or withholding of the requested relief, the court concludes that a permanent injunction must issue. As explained in the Clarification Order, Congress requires the court to permanently enjoin the Navy from obligating or expending any funds for construction of the Everett Homeport until such time as the government fully addresses the environmental concerns, the NEPA violations and the CWA violations, as explained in this memorandum opinion. Clarification Order, supra.

. Although FOE also named the EPA as a defendant, FOE’s motion for summary judgment addressed only the failings of the Corps and the Navy. The EPA has moved to dismiss, and the court will address EPA’s motion in a subsequent order.

. In a subsequent order, the court clarified the scope of its injunction, making it clear that Congress required the court to enjoin the Navy from expending or obligating any funds for the Homeport until the Navy and the Corps fully addressed the environmental issues set out in this memorandum opinion. FOE v. Hall, NO. C88-380R, Order Granting Plaintiffs’ Motion to Clarify (August 8, 1988).

.Wherever appropriate, the facts reflect the Ninth Circuit’s factual statement in Friends of the Earth v. U.S. Navy, 841 F.2d 927, 928-31 (9th Cir.1988). There, the Ninth Circuit enjoined the expenditure of funds on homeport construction until the Navy obtained a permit required by Washington’s Shoreline Management Act. Id. at 937.

. 33 U.S.C. § 1344. A separate permit was required under Section 13 of the Rivers and Harbors Act of 1899, 33 U.S.C. § 407.

. 33 U.S.C. § 1341 (1986).

. The technical appendices contain a number of technical reports on which the agencies relied, including the Corps’ Draft EIS.

. Department of Ecology v. Kirkland, 84 Wash.2d 25, 523 P.2d 1181 (1974).

. This evidence includes the following: (1) FWS and NMFS documents regarding the water quality effects of the Navy’s dredging plans and the Corps’ decision to approve the 404 permit, documents the state boards excluded as hearsay, see Declaration of Todd D. True, exhs. 2-4; and (2) two studies not available at the time of the state board hearings: Palermo, et. al, Evaluation of Dredged. Material Disposal Alternatives for U.S. Navy Homeport at Everett, Washington, Tech. Rep. EL-88, U.S. Army Engineer Waterways Experiment Station, (February, 1988), a report acquired by FOE on May 27, 1988, through discovery in this action; and PTI & Tetra Tech, Everett Harbor Action Program: Analysis of Tox *921 ic Problem Areas (June, 1988) (Draft Report), a study commissioned by EPA.

. See Oklahoma Wildlife Fed. v. U.S. Army Corps of Engineers, 681 F.Supp. 1470, 1487 (N.D.Okl. 1988) (citing Hintz for proposition that 33 C.F.R. 320.4(d) precludes challenge to water quality effects where plaintiff failed to contest state WQC below).

. At one point in the technical appendices, the Corps recognizes that the RADCAD plan, like any other disposal of contaminated material, presents "potential risks to the environment in the event of error and/or failure." FSEIS, App. B, at 15. But the technical appendix, like the EIS’ text, did not reveal the nature and extent of the potential harm to the environment.

. During oral argument, the government cited several decisions similar to Half Moon Bay, in that in each case the reviewing court declined to remand for further agency action. Based on a review of these cases and the rationale underlying each, the court remains convinced that the failure of the ROD to alleviate the areas of environmental concern dictates a remand for further action. See, e.g., Friends of the River v. F.E.R.C., 720 F.2d 93, 106-07 (D.C. Cir.1983) (after finding NEPA violations, the court rejected returning case to agency because subsequent agency actions addressed the problem).

. FOE supports its position with declarations submitted by Dr. Donald C. Malins a well known expert in the field of aquatic toxicology, currently serving as Head of the Environmental Biochemistry Program at the Pacific Northwest Research Foundation and Editor-in-Chief of Aquatic Toxicology, an international journal. While Battelle conducted its tests and the Navy and Corps prepared their EISs, Dr. Malins served as Director of NMFS’ Environmental Conservation Division of the Northwest and Alaska Fisheries Center, and his criticisms were known at that time. See, e.g., FEIS, App.P, ex. 1. In response, the government relies on the declarations of two scientists: (1) Dr. Eric A; Crecelius, a research scientist at Battelle since 1974 and a participant in all three Battelle study phases; and (2) Dr. Jack W. Anderson, currently the Director of the Southern Coastal Water Research Project, formerly senior research scientist at Battelle who participated in Battelle I and II and has reviewed the results of Battelle III.

. By comparing maps found in the FSEIS (at 11-27), the Tetra Tech/PTI Report (at 35), and the Navy’s Final Monitoring Plan (Compl. exh. 12, at 2-2), the court determined that Station EW-11 corresponded to the berm material area and stations EW-5, EW-8, and EW-9 corresponded to contaminated dredging areas. Each of these station sites contained detectable concentrations of a wide variety of the toxic polar compounds at issue. For example, the berm area, EW-11, had detectable concentrations of various resin acids. See Tetra Tech/PTI, Figure 19 at 117, Figure 20 at 119, and Figure 21 at 120. Station EW-11 also had high concentrations of phenol. Figure 25, at 127. Station EW-9 showed the second highest concentration for one type of guaiacol, Figure 28, at 132, while Station EW-5 had the second highest concentration of pentachlorophenol. Figure 29, at 133.

. The NMFS concurred with the decision not to test for certain pulp mill chemicals because “[it] learned that analytical procedures were not sufficiently developed.” Letter of October 5, 1987, at 2 (Compl. exh. 7). Absence of the means to obtain needed information does not release the Corps from its NEPA responsibilities; therefore, a worst-case analysis would have been called for. In any event, the NMFS and the FWS withdrew their recommendation for broader chemical analysis months before publication of the Tetra Tech/PTI Report.

. The Corps informs the court that it disagrees with the Ninth Circuit’s view on the need for a worst-case analysis. In 1979, the analysis requirement became a CEQ regulation codified at 40 C.F.R. § 1502.22, but the CEQ rescinded its regulation in 1986. Methow Valley, 833 F.2d at 817 n. 11. Nevertheless, the Ninth Circuit has held that the rescission does not nullify the requirement because the regulation was merely a codification of prior NEPA case law. Id.; ONRC v. Marsh, 832 F.2d at 1497-98. Even though the Supreme Court granted certiorari on the Corps’ petition challenging the Ninth Circuit’s view, see — U.S.-, 108 S.Ct. 2869, 101 L.Ed.2d 905 (1988), Ninth Circuit precedent controls until such time as the Supreme Court rules otherwise.

. The government also contends that FOE cannot argue that a worst-case analysis is required because it failed to do so at the administrative level. The government bases this argument on the Supreme Court’s decision in Vermont Yankee v. NRDC, 435 U.S. 519, 553-54, 98 S.Ct. 1197, 1216-17, 55 L.Ed.2d 460 (1978). The court has examined the cited pages and found no support for the Corps’ position. Nor do Ninth Circuit cases offer any support for this position. See, e.g., Methow Valley, 833 F.2d at 817; ONRC v. Marsh, 832 F.2d at 1497-98.

. The Corps offers evidence that recent studies indicate only a 10-15% settling rate, resulting in a cap 8 feet thick. Fuglevand Decl. ¶ 12. However, this information is outside the administrative record and not offered for a permissible purpose. See Animal Defense Council v. Hodel, 840 F.2d at 1436; Friends of the Earth v. Hintz, 800 F.2d at 828-29.

. Although Dr. Malins criticizes the use of tracer chemicals rather than actual chemical contaminants, the Corps contends that tracer chemicals were chosen because they are highly soluble and demonstrate a worst-case scenario. Germano Decl., at 18.

.Here again, the Corps attempts to improperly supplement the administrative record with post hoc scientific studies allegedly showing that migration of similar compounds in similar sediments is not taking place in Puget Sound. Cre-celius Decl., at 19-20.

. Dr. Thomas Suchanek, an expert in bioturbation at the Bodega Marine Laboratory, Bodega Bay California, calculated that the Dinnel Survey sampled “an average of roughly one-millionth of the proposed dumping site.” Sucha-nek Decl. ¶ 28.

. See infra pp. 940-942.

. See infra pp. 943, 947.

. Even though the ROD and the Dinnel Study acknowledge that the magnitude of the risks posed by Axiopsis and Molpadia was unknown, the government now claims to have evidence provided by Battelle scientists that Axiopsis and Molpadia "will not pose any threat." Opp. Mem., at 37. Because this evidence is outside the administrative record and does not come within any of the recognized exceptions, the court cannot consider it. However, when the Corps conducts its supplemental EIS, the court suggests that careful attention be paid to Dr. Suchanek’s contention that the "newly acquired” evidence may not be very meaningful. See Su-chanek 2d Decl.

. As noted earlier, the Supreme Court has granted certiorari with respect to Methow Valley. The Corps intends to argue to the Court that neither NEPA nor CEQ regulations contain a requirement to develop a full mitigation plan as part of an EIS. Nevertheless, until the Supreme Court rules otherwise, the Ninth Circuit’s view on the importance of mitigation plans controls in the instant situation.

.Examples of Corps reliance on the monitoring plan are legion, and no purpose would be served in setting out every instance. FOE sets out nine examples. See Reply Memo, at 25 n. 21. Moreover, the government argued to the court that the monitoring plan will “ensure that the environment is protected,” Opp.Mem., at 39 (emphasis added) and that the plan negates the need for preparation of a worst-case analysis. Id. at 52.

. As the court concludes infra, the treatment of the monitoring plan process violated the Corps’ own regulations. See infra p. 947.

. The government contends that the Navy’s final plan eliminated many of the objections of the resource agencies. However, the government failed to specify what objections were eliminated, and the page cited in the administrative record indicates that the Navy claimed to have incorporated the Corps’ comments, but does not indicate anything with respect to the resource agencies. See AR: 44-1.

. As noted earlier the Corps now offers extra-record evidence that the consolidation rate will be 10-15 percent. Because the evidence does not meet any of the exceptions for reviewing extra-record evidence, the court can not consider it.

. See infra, at 946-947.

. The court presents this summary to aid in the Corps' supplementation of the record. However, the summary is not intended to take the place of the detailed discussion and identification of NEPA violations set forth supra.

. For example, from mass release, sediments less than 3 cm thick, and the 5 per cent of contaminants greater than 3 cm thick that need not be capped.

. Congress has appropriated $272 million for Initial Operating Capability ("IOC”). See GAO, Navy Ships: Information on Benefits and Costs 24, 26, 83 (June 1988). The court recognizes that the GAO report concludes that the Home-port will actually cost between $377 and $441 million to house the entire projected battle-group. Id. Nevertheless, the $272 million figure reflects the amount actually budgeted by Congress for IOC, and the Corps was not arbitrary in using this amount.

. Although the FSEIS indicates that after disposal is completed Smith Island retains some economic development potential, FSEIS, App. C, at 11, 13, the Corps subsequently concluded that due to the presence of the disposal site, the need for restrictive covenants in a deed, etc., it was not realistic to assume that a future sale of the island would recoup substantial sums of money. AR: 36-14. Although this information first appeared in a section of the government's supplemental brief that the court struck, the court erred in striking this particular section.

. The FSEIS estimates that the least expensive Smith Island alternative, the excavated option, exceeds RADCAD's $17.5 million cost by anywhere from $16-32 million. FSEIS, III — 104.

. The NMFS estimates $3 million for upland disposal monitoring, but does not provide any specifics on its cost estimate. See NMFS Letter of April 3, 1987 (Compl. exh. 4).

. Although the Corps correctly notes that the court in National Wildlife also relied on the APA, the opinion clearly discussed and focused upon the Corps own public comment requirements. 568 F.Supp. at 993.

. See supra, at 939-940.

. See supra, at 938.

.See supra, at 939-940.